ATAR

atar

Dive into the transformative potential of Augmented Reality (AR) within industrial design. This insightful examination dissects AR's promising impact on traditional prototyping, addressing both the innovative opportunities and inherent biases and underscores our collective responsibility to navigate this new frontier judiciously. June 21, 2023 "The only way to predict the future is to build it." - Alan Kay, computer scientist. June 16, 2023 As technology continues to revolutionize our lives, stands at the forefront, particularly in the realm of . AR has transcended its roots as a novelty for tech enthusiasts and gamers to become a potentially transformative force in our industry. Augmented Reality (AR) industrial design The question is - will AR replace traditional prototyping in industrial design, and if so, at what cost? ‍ AR and Industrial Design: A New Horizon The landscape of industrial design continually evolves. Today, AR promises to ignite another paradigm shift, serving as a medium for designing, iterating, and presenting products, much like how shook the conventions of foam core modeling years ago. 3D printing Recognizing the Power and Potential Biases of AR AR holds considerable promise as an alternative to traditional prototyping, offering the ability to design and modify in a digital 3D space in real-time. However, the adoption of AR as the standard medium for design evaluation might introduce inherent biases. It's crucial for us, as designers, to recognize and account for these potential biases to ensure objective and accurate design evaluations. The Medium Matters: The Impact of AR on Product Evaluation A recent study highlighted in the article reveals that the medium used for product presentation significantly influences user perception. "Can Augmented Reality Supplant Prototyping in Industrial Design?" This influence becomes particularly pronounced when dealing with products with complex geometrical shapes. The medium isn't just a neutral tool but plays a crucial role in shaping user experience and perception. As immersive and impactful AR and VR are in presenting product attributes, they may also skew perceptions and decisions differently than physical prototypes. Mitigating Bias: The Responsibility of Designers in the AR Era These potential biases introduce an important consideration for the future of product design and evaluation. The rise of AR and VR technologies should not be seen as a threat but rather an exciting opportunity to revolutionize our field. As stewards of this evolving landscape, we have a responsibility to acknowledge these biases consciously and adopt careful strategies to counteract them. Involving end-users early in the design process, fostering designer-user collaboration, and offering multiple design options for evaluation are just some ways to mitigate the potential biases of AR. The challenges presented by AR are not barriers but invitations to innovate and evolve our practices. Embracing the Augmented Future: A Balanced Approach While AR holds tremendous promise, it's essential to strike a balance and not abandon traditional prototyping completely. The lack of haptic feedback in AR, the anxiety related to technology adoption, and user familiarity with AR interfaces are all important factors to consider. Our role is to act as mediators between the technology and users, taking a balanced approach to leverage the strengths of AR while also addressing its limitations. Toward a Thoughtful, Bias-Aware Future in Industrial Design The future we envision is one where AR acts as a to the physical world of prototypes, reshaping our conception, development, and experience of products. It's an exciting prospect, but one that requires us to act with awareness and thoughtfulness. digital twin As we move towards this future, it's imperative to realize the potential risks and biases of technology. At Speck Design, we continuously push the boundaries of what's possible, always striving to deliver the best end products. Our commitment to innovation includes recognizing and managing the challenges presented by technologies like AR, ensuring a thoughtful, bias-aware approach to our work. As we step into this exciting new era, let's do so with eyes wide open, acknowledging both the immense opportunities and the important responsibilities it brings. Speck Design Product Design Agency San Jose, CA ‍ At Speck Design, our dedication lies in shaping the future through human-centric industrial design and engineering. Our 25-year journey has been driven by a commitment to bringing memorable experiences to life. In our continuous quest for innovation, we recognize the transformative power of technologies like AR, VR, XR, and MR, but also understand the potential risks and biases they could introduce. As we navigate this exciting journey, we are committed to thoughtfully integrating these technologies, ensuring they serve our ultimate goal - creating remarkable products that truly resonate with users. Let's venture into this augmented future together, conscious of our shared responsibility to shape a future that is not only technologically advanced but also ethical and inclusive. Palacios-Ibáñez, Almudena, et al. "On the Application of Extended Reality Technologies for the Evaluation of Product Characteristics During the Initial Stages of the Product Development Process." Instituto Universitario de Investigación en Tecnología Centrada en el Ser Humano, Universitat Politècnica de València, Ciudad Politécnica de la Innovación, vol. 23, no. 9, 2022, ScienceDirect, doi: 10.1016/S0166361522001762. Aftab, M., & Rusli, H.A. (2017). Designing visceral, behavioural and reflective products. Chin. J. Mech. Eng. (Engl. Ed.), 30 (5), pp. 1058-1068. DOI: 10.1007/s10033-017-0161-x Achiche, S., Maier, A., Milanova, K., & Vadean, A. (2014). Designing visceral, behavioural and reflective products. Chin. J. Mech. Eng. (Engl. Ed.), 11, pp. 1-10. DOI: 10.1115/IMECE2014-40443 Agost, M.-J., Vergara, M., & Bayarri, V. (2021). The use of new presentation technologies in electronic sales environments and their influence on product perception. In: Proceedings of the International Conference on Human-Computer Interaction, 12765, pp. 3-15. DOI: 10.1007/978-3-030-78321-1 Ant Ozok, A., & Komlodi, A. (2009). Better in 3D? An empirical investigation of user satisfaction and preferences concerning two-dimensional and three-dimensional product representations in business-to-consumer e-commerce. Int. J. Hum. Comput. Interact., 25 (4), pp. 243-281. DOI: 10.1080/10447310802546724 Arbeláez, J.C., & Osorio-Gómez, G. (2018). Crowdsourcing augmented reality environment (CARE) for aesthetic evaluation of products in conceptual stage. Comput. Ind., 99, pp. 241-252. DOI: 10.1016/j.compind.2018.03.028 Artacho-Ramírez, M.A., Diego-Mas, J.A., & Alcaide-Marzal, J. (2008). Influence of the mode of graphical representation on the perception of product aesthetic and emotional features: an exploratory study. Int. J. Ind. Ergon., 38 (11–12), pp. 942-952. DOI: 10.1016/j.ergon.2008.02.020 ‍ © 2023 Speck Design. All Rights Reserved

The Augmented Reality Of Product Design: Looking at AR's Role and What It's Hiding

La realidad aumentada del diseño de productos: mirando el papel de AR y lo que está ocultando

La réalité augmentée de la conception de produits : examiner le rôle de la réalité augmentée et ce qu'elle cache

उत्पाद डिज़ाइन की संवर्धित वास्तविकता: एआर की भूमिका और यह क्या छिपा रहा है, इसे देखते हुए

プロダクト デザインの拡張現実: AR の役割とそこに隠されているものを考察する

A realidade aumentada do design de produtos: observando o papel da AR e o que ela está escondendo

Thực tế mở rộng của thiết kế sản phẩm: Nhìn vào vai trò của AR và những gì nó đang che giấu

产品设计的增强现实：了解 AR 的作用及其隐藏的内容

is quite possibly one of the most popular videogames ever made, originally created by Alexey Leonidovich Pajitnov in 1984 and was ported over to a multitude of home computers and consoles. Tetris Every machine that could run the game had its own version made for it, most notably the Nintendo Game Boy (which came bundled with a free copy of the game, being a launch title for the handheld console). The Nintendo Entertainment System actually got two (2) versions; the first one was withdrawn due to licensing issues thanks to a court dispute between Atari and Nintendo, and was replaced with a second, often considered inferior, edition. I am going to take a look at the version that I consider one of the best (official) releases for home computers on the Atari ST, a home computer introduced in 1985 by Atari Corporation and intended to compete with the Commodore Amiga. ’s Legacy Tetris The game’s legacy can be attributed to its simplicity and addictiveness, in addition to unintentionally stimulating the player’s brain, indirectly making them more skilled at certain tasks outside of this game. It is believed that players who have played consistently and for long enough may develop some cognitive advantages such as improved memory, enhanced concentration and increased perceptual awareness. However, this theory has yet to be confirmed or disconfirmed. It can also act as a tranquilliser for people with anxiety, soothing their worries while they play what appears to be a deceptively plain-looking game. Tetris The Game When the game boots up, the player is greeted with a pleasant chiptune composition courtesy of composer David Whittaker along with some opening credits followed by a basic high score table. On this screen, the player can change the difficulty level by pressing and on the numerical keypad to lower and raise it respectively (the lowest possible level is 0, while the highest is 9). When they are ready, they can press the on the keyboard or the fire button on their 9-pin Atari joystick, commonplace to home computers from this era. I will play the game (and recommend playing) with a joystick, as the controls for the keyboard are extremely counterproductive and unintuitive (and cannot be redefined unlike most games from this era). 7 9 Space Bar The goal of the game is to keep the main playing area as clear as possible of falling objects known as , building blocks consisting of four (4) squares connected together orthogonally. This is accomplished by forming a constant line of tetrominoes horizontally (without gaps). tetrominoes When a line is formed, it will disappear, clearing that line of the playing field - additional points can be gained by clearing multiple lines simultaneously. The player is given a set of seven (7) tetrominoes, all named after letters of the Latin alphabet: They are all colour-coded, depending on the version (except for the Game Boy, which uses different textures due to its monochrome display). I, J, L, O, S, T and Z. The player can manipulate the tetrominoes by moving them left and right as they gradually descend from the top of the “sky” in addition to rotating them in either direction. They can accelerate each tetromino’s descent if they desire, although this is usually only recommended if they are absolutely sure of their choice and confident that they can make a line which can be cleared with future tetrominoes. When a certain amount of lines have been cleared, the game’s level of difficulty increases and the game speeds up slightly, getting faster and harder with each successive level. As in most versions, blocks do not automatically fall into open gaps when lines are cleared. The player can see what the next tetromino block will be in the “Next” box (can be toggled on and off on the Atari ST with the key on the numerical keypad), which will begin its descent when the current block has landed and locked into position (the player can still move and rotate the current tetromino until then, providing they have enough space to manoeuvre it into place). 1 Due to its simplistic arcade-like nature (although no arcade version has ever been made officially), the game can only end in a single manner: when the player inevitably fills up the playing field, so the overall goal simply involves getting as high a score as possible (some versions reward the player with an ending cutscene upon reaching a high enough score). Presentation Being a 16-bit version, one would expect more flair than its 8-bit counterparts, and it delivers fairly well, although not exceptionally well due to it being an early title for the Atari ST. A main menu would have been nice as opposed to a basic high score table, although I do like the pseudo-cinematic opening credits. Tetris (Atari ST) Graphics (Detail, Colour) Review Graphically, this game is OK for a 16-bit computer, but it does very little to push the visual envelope, making it look like an enhanced 8-bit title. It is nice that the tetrominoes are all different colours to help differentiate them, but a part of me wishes that the golden psychedelic background in the main playing area could be disabled, as I did find it slightly distracting while playing this game. Most of the visual appeal comes from the title screen. Sound (SFX, Music) The game uses the Atari ST’s YM2149F sound chip. It is similar, but not identical, to the AY-3-8912 chip found in the Amstrad CPC, MSX series of computers and ZX Spectrum machines (starting from the original 128K “Toastrack” model onwards), so the music is going to sound very close between the three main 8-bit computers and this 16-bit machine. There are only three (3) compositions: The main title theme, which plays during the title screen and opening sequence An in-game melody unique to this version A short “game over” fanfare There are no sound effects to speak of and from what I understand, the music cannot be turned off, making this a rare case where the only sound available is music, which is unusual for a game from this era. The lack of sound is disappointing. Controls As stated earlier, the game can be played with either a keyboard or a joystick, but the controls for the keyboard are ridiculously awkward and unconventional, forcing the player to use the numerical keypad, and even then, it takes a long time to adapt, although this is mitigated slightly by the on-screen display telling the player which key does what. Joystick heavily recommended. The controls are also inconsistent in their responsiveness - sometimes a tetromino will rotate or fall at the push of a button, while other times it will require multiple presses to perform a single action. I suspect this may be due to most of the CPU’s runtime being used to generate the psychedelic effect in the background and, unlike the Amiga, the ST has no dedicated graphical assistance, so everything must be done via the processor. Tetris (Atari ST) Gameplay Review When the game runs smoothly, it truly shines and shows the player what the ST can do - it can be fun, frustrating and even taxing, forcing the player to think on their feet and adapt to their new situation, never letting up (there is no way to pause the game as far as I can tell, unlike most versions). However, the gameplay itself is limited, as other than racking up a high score, there is precious little to offer. Overall While on the Atari ST may not be the best overall version of the game (that honour would go to the Nintendo Game Boy), it is the best version officially available for home computers of the era (the Amiga version doesn’t count because it is a straight ST port, which was typical until 1989 with the release of , after which ST-to-Amiga ports became less commonplace). Tetris Shadow of the Beast The gameplay is hindered by the awkward keyboard layout, unnecessary psychedelic visuals and often-unresponsive controls - the real standout in this game is the music. If this is the only version available to you, then by all means, play and enjoy, but you’re better off searching for other versions such as the aforementioned-and-often-compared Game Boy release, or even on the ZX Spectrum, an unofficial clone with some unique gameplay elements of its own such as 2-player competitive play and a distinctly Russian (or rather, Soviet) theme, hailing from the former Czechoslovakia. Tetris 2 Final Score: 3 out of 5.

Tetris (Atari ST) Retro Game Review

Everyone is talking about the race between Artificial Intelligence and Human Intelligence. When will AI fully surpass human ability and be in control of a majority of our daily lives? While humans spend their days going to school and educating themselves, what is AI doing to get an edge on the competition? AI needs to step up it’s game! Jokes apart, there is a technique in Artificial Intelligence called Reinforcement Learning (RL) which allows an AI to train/teach itself how to perform a certain task. In my case, I used RL to train an AI to play the Atari version of Space Invaders! In this article I will go through the concepts involved in creating an RL agent such as the one I made. I will also include my code if you want to see a Python implementation! Before we get started… I will use the following terms throughout the article. If you aren’t familiar with them already, well…here you go! The AI player. In this case it will be the shooter at the bottom attacking the aliens agent — The complete surroundings of the agent (i.e. the barriers in front of it and the aliens above) environment — Something the agent has the option of doing (i.e. move left, move right, shoot, do nothing) action — Choosing and performing 1 action step — The current situation the AI is in state — So how does Reinforcement Learning work? I briefly mentioned that Reinforcement Learning can be used to help an AI train itself, but how does it actually work? In RL, the AI trains on a reward based system. Think back to elementary school. Every time you did something right, your teacher might give you a cookie or a sticker. This would encourage you to do more similar actions because you want as many cookies and stickers as you can get. Every time you did something wrong you might get a timeout or lose recess, which would discourage you from doing that again. The agent learns in a very similar way. We set conditions that if the AI follows correctly, they will receive an imaginary reward. If they do the wrong things they lose these rewards. When we apply this to Space Invaders, the agent extracts observations from the environment which will help it decide the right action to take. For example, if it sees a bullet coming towards it, it will move out of the way! How can our AI understand the environment? Taking a step back, how does our AI “observe” the environment in the first place? How is it able to recognize that a bullet is coming towards it or even what a bullet is? It is able to do all of these things with the use of a Neural Network. Neural Network Diagram Neural Networks (NN) are modeled after the human brain. Like the brain has neurons which transmit information between themselves, a NN has “nodes” (the white circles) which process information and perform operations. In a standard NN, there are three basic types of layers: an input layer, hidden layers, and an output layer. The input layer takes the input, which is processed by the hidden layers, and finally spit out the output in the output layer. Convolutional Neural Network Diagram We will be using a special type of Neural Network known as a Convolutional Neural Network (CNN). What makes CNNs unique is their ability to analyze imagery. They contain convolutional layers which take images as inputs and sweep over them pixel by pixel. Each convolutional layer is designed like a filter, searching for something specific in the image. You can have layers searching for something as general as a square to something more complicated like a bird! Okay, so our AI can tell what is what in our environment. AGENT LOOK OUT A BULLET! Static state Which way is it going? How fast is it going? Will we be able to go across without getting hit? All of these things can’t be determined with a static image, so just like we don’t have answers to these questions, neither does the AI. The last thing we need to do so that our AI can properly understand the environment is images to add directional sense. If we take 1 image, wait a couple frames, take another image, and so on, we are able to create something similar to a GIF for our AI to perceive. stack 4 states stacked on top of each other It’s a lot easier to see that the bullets are coming towards us and at what speed! Learning to play the game with Q-Learning At this point we are giving the AI a reward for shooting the Aliens, the AI knows how to understand the environment, but there is still one fundamental component we are missing. How does the AI know what to do in what situation? If you or I were to play the game, we would have a general idea of what to do. Shoot the aliens and dodge the bullets, simple! But how did we know to do this? From past experiences, we know that getting shot isn’t good and to win a game you probably have to kill all the enemies. The problem is, the AI has no past experiences. Think of it like an infant who has the ability to play video games. So what better way to learn how to play a game then by playing the game! Before we do that, it would be a good idea to have a way of remembering what worked well and what didn’t. So the AI will first create something called a Q-Table. Example Q-Table A Q-Table contains a column for every possible action in your game, and a row for every possible state in your game. In the cell where they meet, we put the we expect by doing that action in the given state. Intuitively, the AI will always perform the action which will render the highest reward. Maximum Future Expected Reward We can’t expect to know all of the future rewards right away because then we wouldn’t need an AI to play the game in the first place. We predict these numbers through the Bellman Equation. If you are like me, you probably got a headache just by looking at it. To break it down, we take the reward from an action/state pair and use a neural network to predict the highest expected reward in the next state. We do this for each possible action and take the highest value. Finally, we perform the action that will yield the calculated reward. Our AI is always looking one step ahead! When the AI actually starts playing the game, we use a training method called At the start, the value of epsilon is very high. While epsilon is high, the AI will choose its actions randomly since there is no knowledge to base its actions upon. As the AI unintentionally stumbles upon rewards or penalties, it starts to associate performing certain actions in certain states as either good or bad and fills in the Q-Table accordingly. As training goes on, the value of epsilon slowly decays. With a lower value for epsilon, there is a higher likelihood that the AI will make decisions based on its Q-Table instead of randomly generating it. epsilon-greedy. By the time epsilon becomes practically 0, the AI should have a well filled out Q-Table to help it make decisions. Let’s look at my AI play Space Invaders With these concepts in mind, I developed my own AI to play Space Invaders. A link to my code is right . here Overall the AI plays pretty well considering the amount of training it did. AIs generally take an EXTREMELY long time to train and a lot of computational power. When you think your AI has learned it all, it stumbles across a new problem! My AI recently discovered that it can lose if the Aliens reach the bottom. As seen in the video, it completely ignored the Aliens which were getting closer and closer. With some more training, it should realize that it can increase its total rewards by shooting the Aliens before they reach the bottom since that will prolong the game. Key Takeaways is a system where the AI is rewarded for doing the right thing thus it learns what to do and what not to do Reinforcement Learning are a special type of Neural Network which can recognize images using Convolutional Neural Networks Convolutional Layers is when the AI plays the game and records what actions in what states give what rewards in a After a lot of training, they used the develop Q-Table to play to the best of their abilities Q-Learning Q-Table.

How I Trained an AI to Play Atari Space Invaders

Table of Links Abstract


Introduction
Background
Reinforcement Learning
Similar Work


Methodology
DRLAgent Design


Training Procedures


Testing Procedures


Results


SABR Experiments


Conclusions


Appendix A


References REFERENCES Assa, H., C. Kenyon, and H. Zhang. 2021. “Assessing Reinforcement Delta Hedging.” SSRN SSRN 3918375. http://dx.doi.org/10.2139/ssrn.3918375. Black, F., and M. Scholes. 1973. “The Pricing of Options and Corporate Liabilities” 81 (3): 637– 54. https://doi.org/10.1086/260062. Cao, Jay, Jacky Chen, Soroush Farghadani, John Hull, Zissis Poulos, Zeyu Wang, and Jun Yuan. 2023. “Gamma and Vega Hedging Using Deep Distributional Reinforcement Learning.” Frontiers in Artificial Intelligence 6. https://www.frontiersin.org/articles/10.3389/frai.2023.1129370. Cao, Jay, Jacky Chen, John Hull, and Zissis Poulos. 2021. “Deep Hedging of Derivatives Using Reinforcement Learning.” The Journal of Financial Data Science 3 (1): 10–27. https://doi.org/10.3905/jfds.2020.1.052. Chebyshev, P. 1864. “Sur l’interpolation.” Zapiski Akademii Nauk 1 (4): 539–60. Du, Jiayi, Muyang Jin, Petter N. Kolm, Gordon Ritter, Yixuan Wang, and Bofei Zhang. 2020. “Deep Reinforcement Learning for Option Replication and Hedging.” The Journal of Financial Data Science 2 (4): 44–57. https://doi.org/10.3905/jfds.2020.1.045. Fathi, A., and B. Hientzsch. 2023. “A Comparison of Reinforcement Learning and Deep Trajectory Based Stochastic Control Agents for Stepwise Mean-Variance Hedging.” arXiv arXiv:2302.07996. Florescu, Ionuţ, and Cristian Gabriel Pãsãricã. 2009. “A Study about the Existence of the Leverage Effect in Stochastic Volatility Models.” Physica A: Statistical Mechanics and Its Applications 388 (4): 419–32. Giurca, B., and S. Borovkova. 2021. “Delta Hedging of Derivatives Using Deep Reinforcement Learning.” SSRN SSRN 3847272. https://ssrn.com/abstract=3847272. Glau, Kathrin, Micro Mahlstedt, and Christian Potz. 2018. “A New Approach for American Option Pricing: The Dynamic Chebyshev Method.” arXiv. https://doi.org/10.48550/arXiv.1806.05579. Hagan, Patrick, Deep Kumar, Andrew Lesniewski, and Diana Woodward. 2002. “Managing Smile Risk.” Wilmott Magazine 1 (January): 84–108. Halperin, Igor. 2017. “QLBS: Q-Learner in the Black-Scholes(-Merton) Worlds.” The Journal of Derivatives 28 (1): 99–122. https://doi.org/10.3905/jod.2020.1.108. Heston, Steven L. 1993. “A Closed-Form Solution for Options with Stochastic Volatility with Applications to Bond and Currency Options.” The Review of Financial Studies 6 (2): 327–43. Hull, John, 1946-. 2012. Options, Futures, and Other Derivatives. Eighth edition. Boston : Prentice Hall, [2012] ©2012. https://search.library.wisc.edu/catalog/9910112878402121. Kolm, Petter N., and Gordon Ritter. 2019. “Dynamic Replication and Hedging: A Reinforcement Learning Approach.” The Journal of Financial Data Science 1 (1): 159–71. https://doi.org/10.3905/jfds.2019.1.1.159. Lillicrap, Timothy, Johnathan J. Hunt, Alexander Pritzel, Nicolas Heess, Tom Erez, Yuval Tassa, David Silver, and Daan Wierstra. 2015. “Continuous Control with Deep Reinforcement Learning.” arXiv arXiv:1509.02971. https://doi.org/10.48550/arXiv.1509.02971. Lin, Long-Ji. 1992. “Self-Improving Reactive Agents Based on Reinforcement Learning, Planning and Teaching.” Machine Learning 8 (3): 293–321. https://doi.org/10.1007/BF00992699. Liu, Peng. 2023. “A Review on Derivative Hedging Using Reinforcement Learning.” The Journal of Financial Data Science. https://doi.org/10.3905/jfds.2023.1.124. Longstaff, Francis A., and Eduardo S. Schwartz. 2001. “Valuing American Options by Simulation: A Simple Least-Squares Approach.” UCLA: Finance. https://escholarship.org/uc/item/43n1k4jb. Mikkilä, Oskari, and Juho Kanniainen. 2023. “Empirical Deep Hedging.” Quantitative Finance 23 (1): 111–22. https://doi.org/10.1080/14697688.2022.2136037. Mnih, V., K. Kavukcuoglu, David Silver, A. Graves, I. Antonoglou, Daan Wierstra, and M. Riedmiller. 2013. “Playing Atari with Deep Reinforcement Learning.” arXiv arXiv:1312.5602. https://doi.org/10.48550/arXiv.1312.5602. Murray, Phillip, B. Wood, H. Buehler, Magnus Wiese, and Mikko Pakkanen. 2022. “Deep Hedging: Continuous Reinforcement Learning for Hedging of General Portfolios across Multiple Risk Aversions.” In ICAIF ’22: Proceedings of the Third ACM International Conference on AI in Finance, 361–68. Association for Computing Machinery. https://doi.org/10.1145/3533271.3561731. Pham, Uyen, Quoc Luu, and Hien Tran. 2021. “Multi-Agent Reinforcement Learning Approach for Hedging Portfolio Problem.” Soft Computing 25 (12): 7877–85. https://doi.org/10.1007/s00500-021-05801-6. Pickard, Reilly, and Yuri Lawryshyn. 2023. “Deep Reinforcement Learning for Dynamic Stock Option Hedging: A Review.” Mathematics 11 (24). https://doi.org/10.3390/math11244943. Silver, David, Aja Huang, Chris J. Maddison, Arthur Guez, Laurent Sifre, George van den Driessche, Julian Schrittwieser, et al. 2016. “Mastering the Game of Go with Deep Neural Networks and Tree Search.” Nature 529 (7587): 484–89. https://doi.org/10.1038/nature16961. Smith, Leslie N. 2018. “A Disciplined Approach to Neural Network Hyper-Parameters: Part 1 -- Learning Rate, Batch Size, Momentum, and Weight Decay.” https://doi.org/10.48550/arXiv.1803.09820. Sutton, Richard S., and Andrew G. Barto. 2018. Reinforcement Learning: An Introduction. A Bradford Book. Watkins, Christopher John Cornish Hellaby. 1989. “Learning from Delayed Rewards.” King’s College. https://www.cs.rhul.ac.uk/~chrisw/new_thesis.pdf. Xiao, B., W. Yao, and X. Zhou. 2021. “Optimal Option Hedging with Policy Gradient.” In 2021 International Conference on Data Mining Workshops (ICDMW), 1112–19. IEEE. https://doi.org/10.1109/ICDMW53433.2021.00145. Xu, Wei, and Bing Dai. 2022. “Delta-Gamma–Like Hedging with Transaction Cost under Reinforcement Learning Technique.” The Journal of Derivatives 29 (5): 60–82. https://doi.org/10.3905/jod.2022.1.156. Zheng, C., J. He, and C. Yang. 2023. “Option Dynamic Hedging Using Reinforcement Learning.” arXiv arXiv:2306.10743. Authors:
(1) Reilly Pickard, Department of Mechanical and Industrial Engineering, University of Toronto, Toronto, ON M5S 3G8, Canada (reilly.pickard@mail.utoronto.ca);
(2) Finn Wredenhagen, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada;
(3) Julio DeJesus, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada;
(4) Mario Schlener, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada;
(5) Yuri Lawryshyn, Department of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, ON M5S 3E5, Canada. This paper is available on arxiv under CC BY-NC-SA 4.0 Deed (Attribution-Noncommercial-Sharelike 4.0 International) license. Table of Links Abstract Introduction
Background
Reinforcement Learning
Similar Work Methodology
DRLAgent Design Training Procedures Testing Procedures Results SABR Experiments Conclusions Appendix A References Abstract Abstract Abstract Introduction Background Reinforcement Learning Similar Work Introduction Background Background Reinforcement Learning Reinforcement Learning Similar Work Similar Work Methodology DRLAgent Design Methodology Methodology DRLAgent Design DRLAgent Design Training Procedures Training Procedures Training Procedures Testing Procedures Testing Procedures Testing Procedures Results Results Results SABR Experiments SABR Experiments SABR Experiments Conclusions Conclusions Conclusions Appendix A Appendix A Appendix A References References References REFERENCES Assa, H., C. Kenyon, and H. Zhang. 2021. “Assessing Reinforcement Delta Hedging.” SSRN SSRN 3918375. http://dx.doi.org/10.2139/ssrn.3918375. Black, F., and M. Scholes. 1973. “The Pricing of Options and Corporate Liabilities” 81 (3): 637– 54. https://doi.org/10.1086/260062. Cao, Jay, Jacky Chen, Soroush Farghadani, John Hull, Zissis Poulos, Zeyu Wang, and Jun Yuan. 2023. “Gamma and Vega Hedging Using Deep Distributional Reinforcement Learning.” Frontiers in Artificial Intelligence 6. https://www.frontiersin.org/articles/10.3389/frai.2023.1129370. Cao, Jay, Jacky Chen, John Hull, and Zissis Poulos. 2021. “Deep Hedging of Derivatives Using Reinforcement Learning.” The Journal of Financial Data Science 3 (1): 10–27. https://doi.org/10.3905/jfds.2020.1.052. Chebyshev, P. 1864. “Sur l’interpolation.” Zapiski Akademii Nauk 1 (4): 539–60. Du, Jiayi, Muyang Jin, Petter N. Kolm, Gordon Ritter, Yixuan Wang, and Bofei Zhang. 2020. “Deep Reinforcement Learning for Option Replication and Hedging.” The Journal of Financial Data Science 2 (4): 44–57. https://doi.org/10.3905/jfds.2020.1.045. Fathi, A., and B. Hientzsch. 2023. “A Comparison of Reinforcement Learning and Deep Trajectory Based Stochastic Control Agents for Stepwise Mean-Variance Hedging.” arXiv arXiv:2302.07996. Florescu, Ionuţ, and Cristian Gabriel Pãsãricã. 2009. “A Study about the Existence of the Leverage Effect in Stochastic Volatility Models.” Physica A: Statistical Mechanics and Its Applications 388 (4): 419–32. Giurca, B., and S. Borovkova. 2021. “Delta Hedging of Derivatives Using Deep Reinforcement Learning.” SSRN SSRN 3847272. https://ssrn.com/abstract=3847272. Glau, Kathrin, Micro Mahlstedt, and Christian Potz. 2018. “A New Approach for American Option Pricing: The Dynamic Chebyshev Method.” arXiv. https://doi.org/10.48550/arXiv.1806.05579. Hagan, Patrick, Deep Kumar, Andrew Lesniewski, and Diana Woodward. 2002. “Managing Smile Risk.” Wilmott Magazine 1 (January): 84–108. Halperin, Igor. 2017. “QLBS: Q-Learner in the Black-Scholes(-Merton) Worlds.” The Journal of Derivatives 28 (1): 99–122. https://doi.org/10.3905/jod.2020.1.108. Heston, Steven L. 1993. “A Closed-Form Solution for Options with Stochastic Volatility with Applications to Bond and Currency Options.” The Review of Financial Studies 6 (2): 327–43. Hull, John, 1946-. 2012. Options, Futures, and Other Derivatives. Eighth edition. Boston : Prentice Hall, [2012] ©2012. https://search.library.wisc.edu/catalog/9910112878402121. Kolm, Petter N., and Gordon Ritter. 2019. “Dynamic Replication and Hedging: A Reinforcement Learning Approach.” The Journal of Financial Data Science 1 (1): 159–71. https://doi.org/10.3905/jfds.2019.1.1.159. Lillicrap, Timothy, Johnathan J. Hunt, Alexander Pritzel, Nicolas Heess, Tom Erez, Yuval Tassa, David Silver, and Daan Wierstra. 2015. “Continuous Control with Deep Reinforcement Learning.” arXiv arXiv:1509.02971. https://doi.org/10.48550/arXiv.1509.02971. Lin, Long-Ji. 1992. “Self-Improving Reactive Agents Based on Reinforcement Learning, Planning and Teaching.” Machine Learning 8 (3): 293–321. https://doi.org/10.1007/BF00992699. Liu, Peng. 2023. “A Review on Derivative Hedging Using Reinforcement Learning.” The Journal of Financial Data Science. https://doi.org/10.3905/jfds.2023.1.124. Longstaff, Francis A., and Eduardo S. Schwartz. 2001. “Valuing American Options by Simulation: A Simple Least-Squares Approach.” UCLA: Finance. https://escholarship.org/uc/item/43n1k4jb. Mikkilä, Oskari, and Juho Kanniainen. 2023. “Empirical Deep Hedging.” Quantitative Finance 23 (1): 111–22. https://doi.org/10.1080/14697688.2022.2136037. Mnih, V., K. Kavukcuoglu, David Silver, A. Graves, I. Antonoglou, Daan Wierstra, and M. Riedmiller. 2013. “Playing Atari with Deep Reinforcement Learning.” arXiv arXiv:1312.5602. https://doi.org/10.48550/arXiv.1312.5602. Murray, Phillip, B. Wood, H. Buehler, Magnus Wiese, and Mikko Pakkanen. 2022. “Deep Hedging: Continuous Reinforcement Learning for Hedging of General Portfolios across Multiple Risk Aversions.” In ICAIF ’22: Proceedings of the Third ACM International Conference on AI in Finance, 361–68. Association for Computing Machinery. https://doi.org/10.1145/3533271.3561731. Pham, Uyen, Quoc Luu, and Hien Tran. 2021. “Multi-Agent Reinforcement Learning Approach for Hedging Portfolio Problem.” Soft Computing 25 (12): 7877–85. https://doi.org/10.1007/s00500-021-05801-6. Pickard, Reilly, and Yuri Lawryshyn. 2023. “Deep Reinforcement Learning for Dynamic Stock Option Hedging: A Review.” Mathematics 11 (24). https://doi.org/10.3390/math11244943. Silver, David, Aja Huang, Chris J. Maddison, Arthur Guez, Laurent Sifre, George van den Driessche, Julian Schrittwieser, et al. 2016. “Mastering the Game of Go with Deep Neural Networks and Tree Search.” Nature 529 (7587): 484–89. https://doi.org/10.1038/nature16961. Smith, Leslie N. 2018. “A Disciplined Approach to Neural Network Hyper-Parameters: Part 1 -- Learning Rate, Batch Size, Momentum, and Weight Decay.” https://doi.org/10.48550/arXiv.1803.09820. Sutton, Richard S., and Andrew G. Barto. 2018. Reinforcement Learning: An Introduction. A Bradford Book. Watkins, Christopher John Cornish Hellaby. 1989. “Learning from Delayed Rewards.” King’s College. https://www.cs.rhul.ac.uk/~chrisw/new_thesis.pdf. Xiao, B., W. Yao, and X. Zhou. 2021. “Optimal Option Hedging with Policy Gradient.” In 2021 International Conference on Data Mining Workshops (ICDMW), 1112–19. IEEE. https://doi.org/10.1109/ICDMW53433.2021.00145. Xu, Wei, and Bing Dai. 2022. “Delta-Gamma–Like Hedging with Transaction Cost under Reinforcement Learning Technique.” The Journal of Derivatives 29 (5): 60–82. https://doi.org/10.3905/jod.2022.1.156. Zheng, C., J. He, and C. Yang. 2023. “Option Dynamic Hedging Using Reinforcement Learning.” arXiv arXiv:2306.10743. Authors: (1) Reilly Pickard, Department of Mechanical and Industrial Engineering, University of Toronto, Toronto, ON M5S 3G8, Canada (reilly.pickard@mail.utoronto.ca); (2) Finn Wredenhagen, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada; (3) Julio DeJesus, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada; (4) Mario Schlener, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada; (5) Yuri Lawryshyn, Department of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, ON M5S 3E5, Canada. Authors: Authors: (1) Reilly Pickard, Department of Mechanical and Industrial Engineering, University of Toronto, Toronto, ON M5S 3G8, Canada ( reilly.pickard@mail.utoronto.ca ); reilly.pickard@mail.utoronto.ca (2) Finn Wredenhagen, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada; (3) Julio DeJesus, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada; (4) Mario Schlener, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada; (5) Yuri Lawryshyn, Department of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, ON M5S 3E5, Canada. This paper is available on arxiv under CC BY-NC-SA 4.0 Deed (Attribution-Noncommercial-Sharelike 4.0 International) license. This paper is available on arxiv under CC BY-NC-SA 4.0 Deed (Attribution-Noncommercial-Sharelike 4.0 International) license. available on arxiv available on arxiv

Hedging American Put Options with Deep Reinforcement Learning: References

Table of Links Abstract


Introduction
Background
Reinforcement Learning
Similar Work


Methodology
DRLAgent Design


Training Procedures


Testing Procedures


Results


SABR Experiments


Conclusions


Appendix A


References REINFORCEMENT LEARNING RL, which is given a thorough explanation by Sutton and Barto (2018), is summarized in this section. RL is a continuous feedback loop between an agent and its environment. At each time step, the RL agent, in a current state 𝑠, takes an action 𝑎, and receives a reward 𝑟. The objective is to maximize future expected rewards, 𝐺𝑡 , which, at time step 𝑡 for an episodic task of 𝑇 discrete steps, are computed as noting that 𝛾 is the discount factor to avoid infinite returns. For any state 𝑠, the corresponding action is given by the agent’s policy, 𝜋(𝑠) → 𝑎. Policies are evaluated by computing value functions such as the 𝑄-function, which maps state-action pairs to their expected returns under the current policy, i.e., SARSA and 𝑄-learning are tabular methods, which require storing the 𝑄-value for each stateaction pair and policy for each state in a lookup table. As a tabular 𝑄-function requires the storage of 𝒮 × 𝒜 values, high-dimensional problems become intractable. However, the 𝑄-function may be approximated with function approximators such as NNs, recalling that this combination of NNs and RL is called DRL. DRL was popularized by the work of Mnih et al. (2013), who approximate the 𝑄-function with a NN, specifically denoting each 𝑄-value by 𝑄(𝑠, 𝑎; 𝜃), where 𝜃 is a vector of the NN parameters that signifies the weights and biases connecting the nodes within the neural network. Mnih et al. (2013) show that this method of deep Q-networks (DQN) can be used to train an agent to master multiple Atari games. The Q-network parameters stored in 𝜃 are optimized by minimizing a loss function between the Q-network output and the target value. The loss function for iteration 𝑖 is given by where 𝜂 is the learning rate for the neural network, and 𝛻𝜃𝑖 𝐿𝑖(𝜃𝑖) is the gradient of the loss with respect to the network weights. In addition to 𝑄-network approximation, another component of DQN is the use of the experience replay buffer (Mnih et al. 2013). Experience replay, originally proposed for RL by Lin (1992), stores encountered transitions (state, action, and reward) in a memory buffer. Mnih et al. (2013) use the replay buffer to uniformly sample the target 𝑄-value for iteration 𝑖 (𝑦𝑖 ). To update the Q-function in value methods such as DQN, a max operation on the 𝑄-value over all next possible actions is required. Further, as discussed above, improving the policy in value methods requires an argmax operation over the entire action-space. When the action-space is continuous, the valuation of each potential next action becomes intractable. In financial option hedging, discretizing the action-space restricts the available hedging decisions. While hedging does require the acquisition of an integer number of shares, a continuous action-space for the optimal hedge provides more accuracy, as the hedge is not limited to a discrete set of choices. As such, continuous action-spaces are much more prevalent in the DRL hedging literature (Pickard and Lawryshyn 2023) Authors:
(1) Reilly Pickard, Department of Mechanical and Industrial Engineering, University of Toronto, Toronto, ON M5S 3G8, Canada (reilly.pickard@mail.utoronto.ca);
(2) Finn Wredenhagen, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada;
(3) Julio DeJesus, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada;
(4) Mario Schlener, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada;
(5) Yuri Lawryshyn, Department of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, ON M5S 3E5, Canada. This paper is available on arxiv under CC BY-NC-SA 4.0 Deed (Attribution-Noncommercial-Sharelike 4.0 International) license. Table of Links Abstract Introduction
Background
Reinforcement Learning
Similar Work Methodology
DRLAgent Design Training Procedures Testing Procedures Results SABR Experiments Conclusions Appendix A References Abstract Abstract Abstract Introduction Background Reinforcement Learning Similar Work Introduction Background Background Reinforcement Learning Reinforcement Learning Similar Work Similar Work Methodology DRLAgent Design Methodology Methodology DRLAgent Design DRLAgent Design Training Procedures Training Procedures Training Procedures Testing Procedures Testing Procedures Testing Procedures Results Results Results SABR Experiments SABR Experiments SABR Experiments Conclusions Conclusions Conclusions Appendix A Appendix A Appendix A References References References REINFORCEMENT LEARNING RL, which is given a thorough explanation by Sutton and Barto (2018), is summarized in this section. RL is a continuous feedback loop between an agent and its environment. At each time step, the RL agent, in a current state 𝑠, takes an action 𝑎, and receives a reward 𝑟. The objective is to maximize future expected rewards, 𝐺𝑡 , which, at time step 𝑡 for an episodic task of 𝑇 discrete steps, are computed as noting that 𝛾 is the discount factor to avoid infinite returns. For any state 𝑠, the corresponding action is given by the agent’s policy, 𝜋(𝑠) → 𝑎. Policies are evaluated by computing value functions such as the 𝑄-function, which maps state-action pairs to their expected returns under the current policy, i.e., SARSA and 𝑄-learning are tabular methods, which require storing the 𝑄-value for each stateaction pair and policy for each state in a lookup table. As a tabular 𝑄-function requires the storage of 𝒮 × 𝒜 values, high-dimensional problems become intractable. However, the 𝑄-function may be approximated with function approximators such as NNs, recalling that this combination of NNs and RL is called DRL. DRL was popularized by the work of Mnih et al. (2013), who approximate the 𝑄-function with a NN, specifically denoting each 𝑄-value by 𝑄(𝑠, 𝑎; 𝜃), where 𝜃 is a vector of the NN parameters that signifies the weights and biases connecting the nodes within the neural network. Mnih et al. (2013) show that this method of deep Q-networks (DQN) can be used to train an agent to master multiple Atari games. The Q-network parameters stored in 𝜃 are optimized by minimizing a loss function between the Q-network output and the target value. The loss function for iteration 𝑖 is given by where 𝜂 is the learning rate for the neural network, and 𝛻𝜃𝑖 𝐿𝑖(𝜃𝑖) is the gradient of the loss with respect to the network weights. In addition to 𝑄-network approximation, another component of DQN is the use of the experience replay buffer (Mnih et al. 2013). Experience replay, originally proposed for RL by Lin (1992), stores encountered transitions (state, action, and reward) in a memory buffer. Mnih et al. (2013) use the replay buffer to uniformly sample the target 𝑄-value for iteration 𝑖 (𝑦𝑖 ). To update the Q-function in value methods such as DQN, a max operation on the 𝑄-value over all next possible actions is required. Further, as discussed above, improving the policy in value methods requires an argmax operation over the entire action-space. When the action-space is continuous, the valuation of each potential next action becomes intractable. In financial option hedging, discretizing the action-space restricts the available hedging decisions. While hedging does require the acquisition of an integer number of shares, a continuous action-space for the optimal hedge provides more accuracy, as the hedge is not limited to a discrete set of choices. As such, continuous action-spaces are much more prevalent in the DRL hedging literature (Pickard and Lawryshyn 2023) Authors: (1) Reilly Pickard, Department of Mechanical and Industrial Engineering, University of Toronto, Toronto, ON M5S 3G8, Canada (reilly.pickard@mail.utoronto.ca); (2) Finn Wredenhagen, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada; (3) Julio DeJesus, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada; (4) Mario Schlener, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada; (5) Yuri Lawryshyn, Department of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, ON M5S 3E5, Canada. Authors: Authors: (1) Reilly Pickard, Department of Mechanical and Industrial Engineering, University of Toronto, Toronto, ON M5S 3G8, Canada ( reilly.pickard@mail.utoronto.ca ); reilly.pickard@mail.utoronto.ca (2) Finn Wredenhagen, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada; (3) Julio DeJesus, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada; (4) Mario Schlener, Ernst & Young LLP, Toronto, ON, M5H 0B3, Canada; (5) Yuri Lawryshyn, Department of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, ON M5S 3E5, Canada. This paper is available on arxiv under CC BY-NC-SA 4.0 Deed (Attribution-Noncommercial-Sharelike 4.0 International) license. This paper is available on arxiv under CC BY-NC-SA 4.0 Deed (Attribution-Noncommercial-Sharelike 4.0 International) license. available on arxiv available on arxiv

How Reinforcement Learning Enhances American Put Option Hedging Strategies

Authors:
(1) Edward Crane, School of Mathematics, University of Bristol, BS8 1TH, UK;
(2) Stanislav Volkov, Centre for Mathematical Sciences, Lund University, Box 118 SE-22100, Lund, Sweden. Table of Links Abstract and Introduction Preliminaries Reduction to the case of uniform geometry All original points are eventually removed, a. s. Proof of Theorem 1 Coupling Y (⋅) and Z(⋅) Acknowledgements and References Appendix Acknowledgment The research of S.V. is partially supported by the Swedish Science Foundation grant VR 2019- 04173 and the Crafoord Foundation grant no. 20190667. S.V. would like to acknowledge the hospitality of the University of Bristol during his visits to Bristol. The research of E.C. is supported by the Heilbronn Institute for Mathematical Research. We would like to thank John Mackay for pointing us to the notion of uniform geometry in [7] and Rami Atar for pointing out the relevant Brownian bees model. Finally, we would like to thank the anonymous referee for many useful comments and recommendations, in particular the suggestion to explain what can be done in the non-convex case. References [1] Addario-Berry, Louigi; Lin, Jessica; Tendron, Thomas. Barycentric Brownian bees, Ann. Appl. Probab. 32 (2022), no. 4, 2504–2539. [2] Grinfeld, Michael; Volkov, Stanislav; Wade, Andrew R. Convergence in a multidimensional randomized Keynesian beauty contest, Adv. in Appl. Probab. 47 (2015), no. 1, 57–82. [3] Durrett, Rick. Probability: theory and examples. Fifth edition. Cambridge University Press (2019). [4] Kennerberg, Philip; Volkov, Stanislav. Jante’s law process, Adv. in Appl. Probab. 50 (2018), no. 2, 414–439. [5] Kennerberg, Philip; Volkov, Stanislav. Convergence in the p-contest, Journal of Statistical Physics 178 (2020), 1096–1125. 6] Kennerberg, Philip; Volkov, Stanislav. A Local Barycentric Version of the Bak-Sneppen Model, Journal of Statistical Physics 182:42 (2021), 17 pp. [7] Leonardi, Gian Paolo; Ritor´e, Manuel; Vernadakis, Efstratios. Isoperimetric Inequalities in Unbounded Convex Bodies. Memoirs of the American Math. Soc., no. 1354, Vol. 276 (2022). [8] Siboni, Maor; Sasorov, Pavel and Meerson, Baruch. Fluctuations of a swarm of Brownian bees, Phys. Rev. E 104 (2021), 7 pp. This paper is available on arxiv under CC 4.0 license. Authors: (1) Edward Crane, School of Mathematics, University of Bristol, BS8 1TH, UK; (2) Stanislav Volkov, Centre for Mathematical Sciences, Lund University, Box 118 SE-22100, Lund, Sweden. Authors: Authors: (1) Edward Crane, School of Mathematics, University of Bristol, BS8 1TH, UK; (2) Stanislav Volkov, Centre for Mathematical Sciences, Lund University, Box 118 SE-22100, Lund, Sweden. Table of Links Abstract and Introduction Abstract and Introduction Preliminaries Preliminaries Reduction to the case of uniform geometry Reduction to the case of uniform geometry All original points are eventually removed, a. s. All original points are eventually removed, a. s. Proof of Theorem 1 Proof of Theorem 1 Coupling Y (⋅) and Z(⋅) Coupling Y (⋅) and Z(⋅) Acknowledgements and References Acknowledgements and References Appendix Appendix Acknowledgment The research of S.V. is partially supported by the Swedish Science Foundation grant VR 2019- 04173 and the Crafoord Foundation grant no. 20190667. S.V. would like to acknowledge the hospitality of the University of Bristol during his visits to Bristol. The research of E.C. is supported by the Heilbronn Institute for Mathematical Research. We would like to thank John Mackay for pointing us to the notion of uniform geometry in [7] and Rami Atar for pointing out the relevant Brownian bees model. Finally, we would like to thank the anonymous referee for many useful comments and recommendations, in particular the suggestion to explain what can be done in the non-convex case. References [1] Addario-Berry, Louigi; Lin, Jessica; Tendron, Thomas. Barycentric Brownian bees, Ann. Appl. Probab. 32 (2022), no. 4, 2504–2539. [2] Grinfeld, Michael; Volkov, Stanislav; Wade, Andrew R. Convergence in a multidimensional randomized Keynesian beauty contest, Adv. in Appl. Probab. 47 (2015), no. 1, 57–82. [3] Durrett, Rick. Probability: theory and examples. Fifth edition. Cambridge University Press (2019). [4] Kennerberg, Philip; Volkov, Stanislav. Jante’s law process, Adv. in Appl. Probab. 50 (2018), no. 2, 414–439. [5] Kennerberg, Philip; Volkov, Stanislav. Convergence in the p-contest, Journal of Statistical Physics 178 (2020), 1096–1125. 6] Kennerberg, Philip; Volkov, Stanislav. A Local Barycentric Version of the Bak-Sneppen Model, Journal of Statistical Physics 182:42 (2021), 17 pp. [7] Leonardi, Gian Paolo; Ritor´e, Manuel; Vernadakis, Efstratios. Isoperimetric Inequalities in Unbounded Convex Bodies. Memoirs of the American Math. Soc., no. 1354, Vol. 276 (2022). [8] Siboni, Maor; Sasorov, Pavel and Meerson, Baruch. Fluctuations of a swarm of Brownian bees, Phys. Rev. E 104 (2021), 7 pp. This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv

Analysis of the Jante’s Law Process and Proof of Conjecture: Acknowledgements and References

The Next Steps for Primordial Black Hole Research

If you've been following the VR industry, you know it's been around for decades. Companies have been trying to make virtual reality a mainstream success for ages. While Apple coined the buzz-worthy term “spatial computing” with their Vision Pro headset, they're certainly not the first to try. From the failed Virtual Boy in the 90s to the moderately successful Oculus and HTC Vive VR headsets, tech giants have struggled to find that elusive killer app. But if you just scroll through TikTok or Instagram, you'll see what's really on people's minds these days. And it definitely isn't about bigger screens to watch Netflix or play games. No, what's driving the conversation is beauty, looks, fashion, and how to glow up. So, as transformative as spatial computing could become, the multi-trillion-dollar global beauty and fashion space will lead AR. Our universal desire to look and feel good provides the real-life use case everyone’s looking for. The promises and challenges of AR beauty The core premise of AR beauty is simple but compelling: Using advanced computer vision and rendering to virtually overlay any makeup look, hairstyle, outfit, or accessory onto your face and body with hyper-realistic precision. Consumers can experiment endlessly, sampling different looks and products with no mess, effort, or commitment. If you've ever ordered a piece of clothing online only for it to look completely different (or just not fit right) when it arrives, you know how frustrating that can be—or bought an expensive watch or pair of kicks, only to regret the purchase when it doesn’t look quite how you expected. What if you could virtually try on anything before committing your money? That's the big idea of AR beauty and fashion—getting rid of the guesswork and potential disappointment by allowing you to realistically experiment with any look or product before pulling the trigger. But, some rudimentary AR makeup apps illustrate the significant technological challenges involved. Poor facial tracking, clunky interfaces, and unconvincing visual realism hamper the experience. Faithfully simulating the intricacies of how cosmetics interact with diverse facial geometries, skin tones, and movements at a high fidelity is an enormously complex computational problem. Implementing AI and AR technologies The latest breakthroughs in AI and AR technologies have proven useful in cracking these challenges. For example, Perfect Corp's AI Makeup Virtual Try-On uses machine learning to realistically render different makeup colors and textures like lipstick, eyeshadow, and more, with the simulations customized to each individual's unique facial features and microexpressions for an ultra-personalized virtual experience. If you're not all that into fashion and beauty, this may not seem like a big deal. But for those obsessed with looking their best, the ability to visualize any look or product before buying could be hugely transformative. Their AI Skin Analysis tech goes further, using artificial intelligence to scan faces and detect 14 different skin conditions and concerns, then surfacing customized product and skincare recommendations tailored to the user. And the AgileHand technology allows realistic virtual try-ons to extend beyond just faces by precisely tracking hand motions. This enables photorealistic AR visualizations for intricate products like jewelry, watches, and nail polish. Some other AR beauty solutions include Makeup AR for virtual makeup try-ons, AI Foundation Shade Finder and Matcher for accurate skin tone detection and foundation shade matching, AI Skin Analysis for accurate skin concern detection and personalized product recommendations of products, and Emulation Technology for Skincare Treatment for accurate representations of the expected results of a course of treatment, overlaid directly onto their faces with the latest advancements in AR technology. The AI tech for AR beauty The recent renaissance of AI breakthroughs, especially around areas like generative adversarial networks (GANs), facial semantic segmentation, and neural rendering, is what's enabling the kind of hyper-realistic and personalized AR experiences that were merely SciFi just years ago. Using GANs, AI models can now learn to generate highly realistic representations of human faces, facial attributes, and facial animations from just a handful of source images. This opens up possibilities for instantly generating virtual makeup looks and hairstyles that are custom-tailored to each user's individual facial geometry and characteristics. Facial semantic segmentation allows AI to precisely understand and map out different facial features like skin, facial hair, eyelashes and more - enabling AR effects to seamlessly integrate and move in harmony with the user's face. Neural rendering techniques leveraging AI can also realistically synthesize the complex ways light interacts with different materials and textures in real time, far beyond what's possible with traditional rendering pipelines. Looking ahead, AI is set to transform AR beauty experiences into something much more akin to having a personal beauty guru. Using deep learning, an AI assistant could continuously analyze your unique attributes, habits, and preferences, automatically surfacing ideal products, trendy looks, and personalized recommendations for you to try virtually in AR with just a voice command or subtle gesture. The possibilities extend far beyond just virtual try-ons. AI could help formulate bespoke beauty products optimized for your individual needs and environmental conditions. Or DTC brands could let customers co-create new cosmetic products and shades with the assistance of AI. The takeaway: Beauty and fashion are the AR killer app While the metaverse is still largely vapourware, the humble makeup tutorial may push spatial computing into the mainstream sooner than we think. With an insatiable consumer appetite for looking our best and plenty of money to be made, the global beauty and fashion industry has both the motivation and resources to drive AR innovation and adoption harder than any other sector. If executed well and key technical barriers around rendering, tracking, and hardware are overcome, AR holds the transformative potential to reshape how we discover, purchase, use, and even perceive beauty products. Immersive virtual try-on experiences, AI-driven personalization, and participatory product development could not only reinvent the beauty consumer experience but also make individual self-expression more accessible than ever before. So, as the tech world covers each new OS update and app ecosystem announcement, keep your eye on the influencers and beauty fanatics pioneering the spatial computing frontier with a fresh face of makeup in every video. They may low-key catalyze the next ubiquitous computing paradigm while just trying to look fly. If you've been following the VR industry, you know it's been around for decades. Companies have been trying to make virtual reality a mainstream success for ages. While Apple coined the buzz-worthy term “ spatial computing ” with their Vision Pro headset, they're certainly not the first to try. spatial computing spatial computing From the failed Virtual Boy in the 90s to the moderately successful Oculus and HTC Vive VR headsets, tech giants have struggled to find that elusive killer app. But if you just scroll through TikTok or Instagram, you'll see what's really on people's minds these days. And it definitely isn't about bigger screens to watch Netflix or play games. No, what's driving the conversation is beauty, looks, fashion, and how to glow up. So, as transformative as spatial computing could become, the multi-trillion-dollar global beauty and fashion space will lead AR. Our universal desire to look and feel good provides the real-life use case everyone’s looking for. The promises and challenges of AR beauty The core premise of AR beauty is simple but compelling: Using advanced computer vision and rendering to virtually overlay any makeup look, hairstyle, outfit, or accessory onto your face and body with hyper-realistic precision. Consumers can experiment endlessly, sampling different looks and products with no mess, effort, or commitment. If you've ever ordered a piece of clothing online only for it to look completely different (or just not fit right) when it arrives, you know how frustrating that can be—or bought an expensive watch or pair of kicks, only to regret the purchase when it doesn’t look quite how you expected. frustrating frustrating What if you could virtually try on anything before committing your money? That's the big idea of AR beauty and fashion—getting rid of the guesswork and potential disappointment by allowing you to realistically experiment with any look or product before pulling the trigger. But, some rudimentary AR makeup apps illustrate the significant technological challenges involved. Poor facial tracking, clunky interfaces, and unconvincing visual realism hamper the experience. Faithfully simulating the intricacies of how cosmetics interact with diverse facial geometries, skin tones, and movements at a high fidelity is an enormously complex computational problem. Implementing AI and AR technologies The latest breakthroughs in AI and AR technologies have proven useful in cracking these challenges. For example, Perfect Corp's AI Makeup Virtual Try-On uses machine learning to realistically render different makeup colors and textures like lipstick, eyeshadow, and more, with the simulations customized to each individual's unique facial features and microexpressions for an ultra-personalized virtual experience. AI Makeup Virtual Try-On If you're not all that into fashion and beauty, this may not seem like a big deal. But for those obsessed with looking their best, the ability to visualize any look or product before buying could be hugely transformative. Their AI Skin Analysis tech goes further, using artificial intelligence to scan faces and detect 14 different skin conditions and concerns, then surfacing customized product and skincare recommendations tailored to the user. AI Skin Analysis And the AgileHand technology allows realistic virtual try-ons to extend beyond just faces by precisely tracking hand motions. This enables photorealistic AR visualizations for intricate products like jewelry, watches, and nail polish. Some other AR beauty solutions include Makeup AR for virtual makeup try-ons, AI Foundation Shade Finder and Matcher for accurate skin tone detection and foundation shade matching, AI Skin Analysis for accurate skin concern detection and personalized product recommendations of products, and Emulation Technology for Skincare Treatment for accurate representations of the expected results of a course of treatment, overlaid directly onto their faces with the latest advancements in AR technology. The AI tech for AR beauty The recent renaissance of AI breakthroughs, especially around areas like generative adversarial networks (GANs), facial semantic segmentation, and neural rendering, is what's enabling the kind of hyper-realistic and personalized AR experiences that were merely SciFi just years ago. Using GANs, AI models can now learn to generate highly realistic representations of human faces, facial attributes, and facial animations from just a handful of source images. This opens up possibilities for instantly generating virtual makeup looks and hairstyles that are custom-tailored to each user's individual facial geometry and characteristics. Facial semantic segmentation allows AI to precisely understand and map out different facial features like skin, facial hair, eyelashes and more - enabling AR effects to seamlessly integrate and move in harmony with the user's face. Neural rendering techniques leveraging AI can also realistically synthesize the complex ways light interacts with different materials and textures in real time, far beyond what's possible with traditional rendering pipelines. Looking ahead, AI is set to transform AR beauty experiences into something much more akin to having a personal beauty guru. Using deep learning, an AI assistant could continuously analyze your unique attributes, habits, and preferences, automatically surfacing ideal products, trendy looks, and personalized recommendations for you to try virtually in AR with just a voice command or subtle gesture. The possibilities extend far beyond just virtual try-ons. AI could help formulate bespoke beauty products optimized for your individual needs and environmental conditions. Or DTC brands could let customers co-create new cosmetic products and shades with the assistance of AI. The takeaway: Beauty and fashion are the AR killer app While the metaverse is still largely vapourware, the humble makeup tutorial may push spatial computing into the mainstream sooner than we think. With an insatiable consumer appetite for looking our best and plenty of money to be made, the global beauty and fashion industry has both the motivation and resources to drive AR innovation and adoption harder than any other sector. If executed well and key technical barriers around rendering, tracking, and hardware are overcome, AR holds the transformative potential to reshape how we discover, purchase, use, and even perceive beauty products. Immersive virtual try-on experiences, AI-driven personalization, and participatory product development could not only reinvent the beauty consumer experience but also make individual self-expression more accessible than ever before. So, as the tech world covers each new OS update and app ecosystem announcement, keep your eye on the influencers and beauty fanatics pioneering the spatial computing frontier with a fresh face of makeup in every video. They may low-key catalyze the next ubiquitous computing paradigm while just trying to look fly.

Why the Beauty and Fashion Industry is the Future of AR and Spatial Computing

কেন সৌন্দর্য এবং ফ্যাশন শিল্প এআর এবং স্থানিক কম্পিউটিং এর ভবিষ্যত

Warum die Beauty- und Modebranche die Zukunft von AR und Spatial Computing ist

Por qué la industria de la belleza y la moda es el futuro de la RA y la computación espacial

Pourquoi l'industrie de la beauté et de la mode est l'avenir de la réalité augmentée et de l'informatique spatiale

सौंदर्य और फैशन उद्योग क्यों AR और स्थानिक कंप्यूटिंग का भविष्य है

美容とファッション業界が AR と空間コンピューティングの未来である理由

뷰티 및 패션 산업이 AR 및 공간 컴퓨팅의 미래인 이유

Por que a indústria da beleza e da moda é o futuro da AR e da computação espacial

Почему индустрия красоты и моды — это будущее дополненной реальности и пространственных вычислений

Güzellik ve Moda Endüstrisi Neden AR ve Uzamsal Bilgi İşlemin Geleceği?

Tại sao ngành làm đẹp và thời trang là tương lai của AR và điện toán không gian

为什么美容和时尚行业是 AR 和空间计算的未来

Introducing Yet Another Umpteenth Programming Language Yikes, gadzooks, and shiver me timbers! Yet another "new" programming language? Yes, Virginia, another new programming language. ZeptoN. Overview So, what the do-while is ZeptoN? Simple, ZeptoN is a simple programming language designed with two objectives... Simple learning language with common syntax style and structure.


A programming language for quick-and-dirty “throwaway programs.” My Programming History As a young boy and a teenager, I cut my teeth learning to program several programming languages. The first was the infamous and ubiquitous BASIC, and the other was Pascal. This was the 1980s, and the home computer revolution was underway. I describe how I got into software here xxx, and getting into programming led me to become a senior software engineer. Yet, one key feature of BASIC and Pascal was simplicity. Perhaps too simple in some ways, but the simplicity allowed the newbie or “noob” coder to start writing, running, and tinkering with code. I had subscriptions to “COMPUTE!” and bought programming books when I went to the mall. Yet, I could type in a program, run the code, and often, play computer games. BASIC Text Adventure I went so far as to write my own “electronic novel” which was a shameless imitation of some of the popular text adventures from Infocom, Adventure International. Mine was a murder mystery that “borrowed” from some other games, including the role-playing ones from companies like SirTech. The games were first written in BASIC, ultimately reaching GW-BASIC, and later in Turbo Pascal. This was more for the fun of creating and to impress friends to see if they could find the clues and solve the mystery. From this fascination with text adventures, I learned compiler theory concepts like lexical analysis, syntax, parsing, BNF, etc. Keep It Simple Silly Yet, there was the simplicity of BASIC and Pascal. Both programming languages were simple enough to teach important programming concepts, and often not formally. Then both programming languages were powerful enough and expressive enough to do neat and cool things, like my own text adventure. Now, when I look at introductory programming courses, powerful programming languages are often used--like Java, C++, and so forth. Great that mainstream languages are utilized that are in demand; there is only one small problem: Java, C++, C#, and Swift languages are very complex as they use the object-oriented paradigm. For a noob or newb, for a newbie coder, that is a challenging and often frustrating language to learn. I initially experienced that at university, when the computer science department switched from a Pascal curriculum to a C++ one. The big frustration was going from a structured to an object-oriented paradigm. Learn to Fly With a Jetliner The simile I use is using Java/C++/C#/Swift/etc. is like learning to fly a plane by learning to pilot a Boeing 747 or an Airbus A330. A powerful aircraft, but overloading in terms of features, controls, and operation. Kernighan [Kerni 1981] even wrote that “Comparing C and Pascal is rather like comparing a Learjet to a Piper Cub — one is meant for getting something done while the other is meant for learning.” Yet, in this case, ZeptoN is meant for learning. ZeptoN has C/C++/Java syntax, so the most ubiquitous syntax style is used and is familiar, and ZeptoN transpiles or transcompiles from ZeptoN into Java. Thus, ZeptoN leverages Java ubiquity and platform neutrality. Hence, a newb can code ZeptoN on the MacOS, Linux, or Windows platforms. Back To BASIC One problem with BASIC (and to a lesser degree Pascal) was that each platform/computer--Atari 800, Commodore 64, IBM PC, Tandy 1000EX had its own “flavor” of BASIC. Thus, one BASIC program for the TI99/4A was not quite compatible with the TRS-80 BASIC, etc. I often found myself “adapting” BASIC programs to the computer I had at the time. ZeptoN avoids this by using Java, and having a standard environment of 70+ functions and several standard constants so ZeptoN can run anywhere. All ZeptoN requires is a favorite JDK vendor, with a minimum of JDK 11 for the transcompiler or code editor, Zeptor. Moreover, ZeptoN is free (yes as in free beer) and open-sourced using the GNU GPL v3. Hence, no cost and no proprietary or secret “black box” tool to transpile ZeptoN programs into Java .class bytecode. Testing Ideas With Throwaway Programs As a senior software engineer with 30+ years of writing code, I do like to write “test” programs to try something, and then throw them away. ZeptoN’s simplicity makes this possible, so for an advanced coder to the noob coder, ZeptoN is for everyone! For more details and examples check out the ZeptoN programming language white paper, a 49-page PDF or ePub document, for more details, and some samples of ZeptoN programs. Further Reading For more details and examples check out the ZeptoN programming language white paper and some samples of ZeptoN programs. More about ZeptoN here: https://github.com/wgilreath/ZeptoN and on my home site: https://www.wfgilreath.xyz/ Also the code editor Zeptor: https://github.com/wgilreath/Zeptor Here's ZeptoN in action and in use on Zeptor on YouTube! Enjoy the zither music... 😉 More articles to follow... 😜 about ZeptoN and Zeptor. Zept to it! References [Kerni 1981] Kernighan, Brian W. "Why Pascal is Not My Favorite Programming Language,"
Computing Science Technical Report No. 100, AT&T Bell Laboratories,
April 2, 1981.
https://www.lysator.liu.se/c/bwk-on-pascal.html
https://archive.org/details/pascal-not-my-favorite/page/n2/mode/1up


[Gilre 2024] Gilreath, William F. "Begin at the Beginning...My Start in Software Engineering," PDF (60K) https://bit.ly/will_soft_start


[Gilre 2022] Gilre 2022, Gilreath, William F. "ZeptoN Programming Language," ePub (5.7Mb) https://wgilreath.github.io/ZeptonWhitePaper.epub


[Gilre 2022] Gilre 2022, Gilreath, William F. "ZeptoN Programming Language," PDF (5.yMb) https://wgilreath.github.io/ZeptonWhitePaper.pdf Introducing Yet Another Umpteenth Programming Language Yikes, gadzooks, and shiver me timbers! Yet another "new" programming language? Yes, Virginia, another new programming language. ZeptoN. Overview So, what the do-while is ZeptoN? Simple, ZeptoN is a simple programming language designed with two objectives... Simple learning language with common syntax style and structure. A programming language for quick-and-dirty “throwaway programs.” Simple learning language with common syntax style and structure. Simple learning language with common syntax style and structure. A programming language for quick-and-dirty “throwaway programs.” A programming language for quick-and-dirty “throwaway programs.” My Programming History As a young boy and a teenager, I cut my teeth learning to program several programming languages. The first was the infamous and ubiquitous BASIC, and the other was Pascal. This was the 1980s, and the home computer revolution was underway. I describe how I got into software here xxx, and getting into programming led me to become a senior software engineer. Yet, one key feature of BASIC and Pascal was simplicity. Perhaps too simple in some ways, but the simplicity allowed the newbie or “noob” coder to start writing, running, and tinkering with code. I had subscriptions to “COMPUTE!” and bought programming books when I went to the mall. Yet, I could type in a program, run the code, and often, play computer games. BASIC Text Adventure I went so far as to write my own “electronic novel” which was a shameless imitation of some of the popular text adventures from Infocom, Adventure International. Mine was a murder mystery that “borrowed” from some other games, including the role-playing ones from companies like SirTech. The games were first written in BASIC, ultimately reaching GW-BASIC, and later in Turbo Pascal. This was more for the fun of creating and to impress friends to see if they could find the clues and solve the mystery. From this fascination with text adventures, I learned compiler theory concepts like lexical analysis, syntax, parsing, BNF, etc. Keep It Simple Silly Yet, there was the simplicity of BASIC and Pascal. Both programming languages were simple enough to teach important programming concepts, and often not formally. Then both programming languages were powerful enough and expressive enough to do neat and cool things, like my own text adventure. Now, when I look at introductory programming courses, powerful programming languages are often used--like Java, C++, and so forth. Great that mainstream languages are utilized that are in demand; there is only one small problem: Java, C++, C#, and Swift languages are very complex as they use the object-oriented paradigm. For a noob or newb, for a newbie coder, that is a challenging and often frustrating language to learn. I initially experienced that at university, when the computer science department switched from a Pascal curriculum to a C++ one. The big frustration was going from a structured to an object-oriented paradigm. Learn to Fly With a Jetliner The simile I use is using Java/C++/C#/Swift/etc. is like learning to fly a plane by learning to pilot a Boeing 747 or an Airbus A330. A powerful aircraft, but overloading in terms of features, controls, and operation. Kernighan [Kerni 1981] even wrote that “Comparing C and Pascal is rather like comparing a Learjet to a Piper Cub — one is meant for getting something done while the other is meant for learning.” Yet, in this case, ZeptoN is meant for learning. ZeptoN has C/C++/Java syntax, so the most ubiquitous syntax style is used and is familiar, and ZeptoN transpiles or transcompiles from ZeptoN into Java. Thus, ZeptoN leverages Java ubiquity and platform neutrality. Hence, a newb can code ZeptoN on the MacOS, Linux, or Windows platforms. Back To BASIC One problem with BASIC (and to a lesser degree Pascal) was that each platform/computer--Atari 800, Commodore 64, IBM PC, Tandy 1000EX had its own “flavor” of BASIC. Thus, one BASIC program for the TI99/4A was not quite compatible with the TRS-80 BASIC, etc. I often found myself “adapting” BASIC programs to the computer I had at the time. ZeptoN avoids this by using Java, and having a standard environment of 70+ functions and several standard constants so ZeptoN can run anywhere. All ZeptoN requires is a favorite JDK vendor, with a minimum of JDK 11 for the transcompiler or code editor, Zeptor. Moreover, ZeptoN is free (yes as in free beer) and open-sourced using the GNU GPL v3. Hence, no cost and no proprietary or secret “black box” tool to transpile ZeptoN programs into Java .class bytecode. Testing Ideas With Throwaway Programs As a senior software engineer with 30+ years of writing code, I do like to write “test” programs to try something, and then throw them away. ZeptoN’s simplicity makes this possible, so for an advanced coder to the noob coder, ZeptoN is for everyone! For more details and examples check out the ZeptoN programming language white paper, a 49-page PDF or ePub document, for more details, and some samples of ZeptoN programs. Further Reading For more details and examples check out the ZeptoN programming language white paper and some samples of ZeptoN programs. More about ZeptoN here: https://github.com/wgilreath/ZeptoN and on my home site: https://www.wfgilreath.xyz/ Also the code editor Zeptor: https://github.com/wgilreath/Zeptor https://github.com/wgilreath/ZeptoN https://www.wfgilreath.xyz/ https://github.com/wgilreath/Zeptor Here's ZeptoN in action and in use on Zeptor on YouTube ! Enjoy the zither music... 😉 YouTube More articles to follow... 😜 about ZeptoN and Zeptor. Zept to it! References [Kerni 1981] Kernighan, Brian W. "Why Pascal is Not My Favorite Programming Language,"
Computing Science Technical Report No. 100, AT&T Bell Laboratories,
April 2, 1981.
https://www.lysator.liu.se/c/bwk-on-pascal.html
https://archive.org/details/pascal-not-my-favorite/page/n2/mode/1up [Gilre 2024] Gilreath, William F. "Begin at the Beginning...My Start in Software Engineering," PDF (60K) https://bit.ly/will_soft_start [Gilre 2022] Gilre 2022, Gilreath, William F. "ZeptoN Programming Language," ePub (5.7Mb) https://wgilreath.github.io/ZeptonWhitePaper.epub [Gilre 2022] Gilre 2022, Gilreath, William F. "ZeptoN Programming Language," PDF (5.yMb) https://wgilreath.github.io/ZeptonWhitePaper.pdf [Kerni 1981] Kernighan, Brian W. "Why Pascal is Not My Favorite Programming Language,"
Computing Science Technical Report No. 100, AT&T Bell Laboratories,
April 2, 1981.
https://www.lysator.liu.se/c/bwk-on-pascal.html
https://archive.org/details/pascal-not-my-favorite/page/n2/mode/1up [Kerni 1981] Kernighan, Brian W. "Why Pascal is Not My Favorite Programming Language,"
Computing Science Technical Report No. 100, AT&T Bell Laboratories,
April 2, 1981. https://www.lysator.liu.se/c/bwk-on-pascal.html https://archive.org/details/pascal-not-my-favorite/page/n2/mode/1up https://www.lysator.liu.se/c/bwk-on-pascal.html https://archive.org/details/pascal-not-my-favorite/page/n2/mode/1up [Gilre 2024] Gilreath, William F. "Begin at the Beginning...My Start in Software Engineering," PDF (60K) https://bit.ly/will_soft_start [Gilre 2024] Gilreath, William F. "Begin at the Beginning...My Start in Software Engineering," PDF (60K) https://bit.ly/will_soft_start https://bit.ly/will_soft_start [Gilre 2022] Gilre 2022, Gilreath, William F. "ZeptoN Programming Language," ePub (5.7Mb) https://wgilreath.github.io/ZeptonWhitePaper.epub [Gilre 2022] Gilre 2022, Gilreath, William F. "ZeptoN Programming Language," ePub (5.7Mb) https://wgilreath.github.io/ZeptonWhitePaper.epub https://wgilreath.github.io/ZeptonWhitePaper.epub [Gilre 2022] Gilre 2022, Gilreath, William F. "ZeptoN Programming Language," PDF (5.yMb) https://wgilreath.github.io/ZeptonWhitePaper.pdf [Gilre 2022] Gilre 2022, Gilreath, William F. "ZeptoN Programming Language," PDF (5.yMb) https://wgilreath.github.io/ZeptonWhitePaper.pdf https://wgilreath.github.io/ZeptonWhitePaper.pdf

What the Do-While Is ZeptoN?

Imagine you want to go for a bike ride. It could be a social ride with friends, a workout to maintain your fitness, or perhaps you’re an elite cyclist aiming to compete with the world’s best. Unfortunately, poor weather, bad road conditions, or time constraints hold you back. So, you take your bike, put it on a specialized equipment known as a smart trainer, connect it to your PC, Apple TV, or smartphone, and launch an application. In just a few seconds, you’re pedaling through a beautiful virtual world alongside friends, experiencing routes you’d never find on real roads. While this might sound like science fiction or even magic to some, it’s a daily routine for millions, and I would like to give you a glimpse of the technology behind it. As a PC/Console game developer and a passionate cyclist, I spent the last two years working on several experimental projects at the intersection of gaming and indoor cycling within the specialized cycling industry. For example, we released a mobile game for cyclists called Hustle City. If any readers have had a chance to play it, I’d be happy to read about your experience in the comments. I’m writing this post to remind myself of how incredible this industry is and to help others discover its magic. Now, let’s dive into the world of gamified cycling! The Journey of Indoor Cycling The Pioneers The first attempt to create gamified indoor cycling came from the video game pioneer Atari. The 80s witnessed the boom of aerobics and video workouts, and video game companies didn’t want to stay behind. So, in 1982, Atari created a prototype of Project Puffer. Here is how Puffer was going to work: a player pedaled the exercise bike, controlling the speed of a character on the screen. The handlebar of the bike had buttons for navigation so that the player could turn and avoid obstacles. The speed control technique was very straightforward: the faster you pedal, the faster your character goes. Games for Puffer didn’t only include bikes but also car racing and even ship navigation. The project had very high hopes and was anticipated by the public. However, Atari was on the verge of bankruptcy due to the game industry’s global crisis and was eventually passed to a new owner. The ambitious project was abandoned, but it remains in the hearts of vintage Atari enthusiasts. The New Beginning The actual rise of gamified indoor cycling occurred only 30 years later, thanks to advances in cycling technology and rising interest in gamified fitness. The pivotal invention that allowed this to happen was the smart trainer. It finally allowed millions of cycling enthusiasts to bring their workouts indoors at a reasonable price. Now, what makes it smart? Basically, it is a stationary indoor bicycle trainer that can also measure the power you produce and dynamically adjust the pedaling difficulty. The first smart trainers with rich user experiences and third-party device compatibility hit the market in the late 2000s. They were created by the hardware companies like Wahoo, Tacx, and Elite. Since then, about 10 million units have been sold. This invention nurtured a niche but large enough market of cycling enthusiasts hungry for more effective and engaging riding experiences. Cycling hardware companies have paved the way for software companies to enter the gamified cycling experience market. Zwift was the pioneer that introduced its open-world MMO in 2014, and it was a great success. Zwift remains the industry leader today, with around 2 million paying users, but many platforms have followed, including MyWhoosh, Virtupro, and Rouvy. Fun fact: now, indoor cycling platforms even become major sponsors of real-world professional cycling events and teams. Challenges on the Road With the growing popularity of gamified workouts, some unsolved problems still keep the industry within the niche of enthusiasts. “Games Are Not for Efficient Workouts” Most cyclists are reluctant to invest in entertainment and don’t want to sacrifice workout efficiency for the sake of fun. At the same time, they often watch Netflix or YouTube while training because they get bored of pedaling in the same spot with the unchanging scenery. It is both a marketing and game design challenge to help gamified cycling overcome the “leisure activity” stigma and develop robust experiences that encourage users to explore an interactive game during their workout. Lack of Interactivity The most significant limitation of current gamified indoor cycling is the lack of interactivity. With the smart trainer as the primary controller, the interactive aspects are mainly limited to physically based movement calculations and the visual representation of avatars. Control in such games is often done by turning the steering wheel to which the phone is attached or using a piece of special steering equipment. This approach has the following downsides: Players have to mount a phone holder on their bike and put something under the front wheel to turn the wheel more comfortably
The handlebar turn control does not feel natural, as in reality, steering only starts by turning the handlebars and continues by tilting the body, and this handlebar turn is only done by a few degrees
Only the turn movement is available because it is too inconvenient to press any buttons on the phone during the game Although new controllers like Zwift Play offer some improvements, they are product-specific and don’t solve the issue universally. Game Availability and Reliability Maintenance announcements in video games are not uncommon, but they are unacceptable in training software. Having to reschedule your workout due to maintenance can be highly annoying. Imagine an athlete who has already put on their cycling kit, filled a water bottle, and set their mind on a training session. Needless to say, the churn rate spikes dramatically whenever the system is unavailable, and it’s even higher if there’s a crash during a workout. Hardware Communication Many companies produce smart trainers and other equipment for indoor cycling, and it’s not uncommon for users to combine different brands in their setup. The problem is that companies sometimes utilize various communication protocols. Even when they claim to follow standard protocols, their implementations may vary. When users may own so many possible combinations of sports equipment, it isn’t easy to ensure reliable communication and seamless compatibility. The Next Big Move After spending some time on the projects in this area, it became evident that additional specialized equipment is required for a truly engaging indoor cycling experience. Ideally, it should be a third-party solution compatible with different equipment, providing different interaction methods. It turned out that such a platform already exists—it’s a VR headset that is almost a perfect fit for indoor cycling. It’s always astonishing when an emerging technology matures so that new applications adopt it. A VR headset not only enables the use of natural movements as input while cycling, but it also comes equipped with all the necessary hardware features for an optimal indoor cycling platform, including: An all-in-one hardware unit, eliminating the need for a separate TV or PC setup


High-quality audio input and output


Engaging visual capabilities


Robust Bluetooth connectivity for external hardware support Of course, some challenges exist, such as reduced ventilation, sweating, and limited battery life. However, based on my experience, these issues are manageable and don’t affect my enthusiasm for the notion that the future of gamified indoor cycling lies in VR. I’d love to hear your thoughts on this! Imagine you want to go for a bike ride. It could be a social ride with friends, a workout to maintain your fitness, or perhaps you’re an elite cyclist aiming to compete with the world’s best. Unfortunately, poor weather, bad road conditions, or time constraints hold you back. So, you take your bike, put it on a specialized equipment known as a smart trainer, connect it to your PC, Apple TV, or smartphone, and launch an application. In just a few seconds, you’re pedaling through a beautiful virtual world alongside friends, experiencing routes you’d never find on real roads. While this might sound like science fiction or even magic to some, it’s a daily routine for millions, and I would like to give you a glimpse of the technology behind it. As a PC/Console game developer and a passionate cyclist, I spent the last two years working on several experimental projects at the intersection of gaming and indoor cycling within the specialized cycling industry. For example, we released a mobile game for cyclists called Hustle City. If any readers have had a chance to play it, I’d be happy to read about your experience in the comments. I’m writing this post to remind myself of how incredible this industry is and to help others discover its magic. Now, let’s dive into the world of gamified cycling! The Journey of Indoor Cycling The Pioneers The first attempt to create gamified indoor cycling came from the video game pioneer Atari. The 80s witnessed the boom of aerobics and video workouts, and video game companies didn’t want to stay behind. So, in 1982, Atari created a prototype of Project Puffer. Here is how Puffer was going to work: a player pedaled the exercise bike, controlling the speed of a character on the screen. The handlebar of the bike had buttons for navigation so that the player could turn and avoid obstacles. The speed control technique was very straightforward: the faster you pedal, the faster your character goes. Games for Puffer didn’t only include bikes but also car racing and even ship navigation. The project had very high hopes and was anticipated by the public. However, Atari was on the verge of bankruptcy due to the game industry’s global crisis and was eventually passed to a new owner. The ambitious project was abandoned, but it remains in the hearts of vintage Atari enthusiasts. The New Beginning The New Beginning The actual rise of gamified indoor cycling occurred only 30 years later, thanks to advances in cycling technology and rising interest in gamified fitness. The pivotal invention that allowed this to happen was the smart trainer. It finally allowed millions of cycling enthusiasts to bring their workouts indoors at a reasonable price. Now, what makes it smart? Basically, it is a stationary indoor bicycle trainer that can also measure the power you produce and dynamically adjust the pedaling difficulty. The first smart trainers with rich user experiences and third-party device compatibility hit the market in the late 2000s. They were created by the hardware companies like Wahoo, Tacx, and Elite. Since then, about 10 million units have been sold. This invention nurtured a niche but large enough market of cycling enthusiasts hungry for more effective and engaging riding experiences. Cycling hardware companies have paved the way for software companies to enter the gamified cycling experience market. Zwift was the pioneer that introduced its open-world MMO in 2014, and it was a great success. Zwift remains the industry leader today, with around 2 million paying users, but many platforms have followed, including MyWhoosh, Virtupro, and Rouvy. Fun fact: now, indoor cycling platforms even become major sponsors of real-world professional cycling events and teams. Challenges on the Road With the growing popularity of gamified workouts, some unsolved problems still keep the industry within the niche of enthusiasts. “Games Are Not for Efficient Workouts” Most cyclists are reluctant to invest in entertainment and don’t want to sacrifice workout efficiency for the sake of fun. At the same time, they often watch Netflix or YouTube while training because they get bored of pedaling in the same spot with the unchanging scenery. Netflix YouTube It is both a marketing and game design challenge to help gamified cycling overcome the “leisure activity” stigma and develop robust experiences that encourage users to explore an interactive game during their workout. Lack of Interactivity The most significant limitation of current gamified indoor cycling is the lack of interactivity. With the smart trainer as the primary controller, the interactive aspects are mainly limited to physically based movement calculations and the visual representation of avatars. Control in such games is often done by turning the steering wheel to which the phone is attached or using a piece of special steering equipment. This approach has the following downsides: This approach has the following downsides: Players have to mount a phone holder on their bike and put something under the front wheel to turn the wheel more comfortably The handlebar turn control does not feel natural, as in reality, steering only starts by turning the handlebars and continues by tilting the body, and this handlebar turn is only done by a few degrees Only the turn movement is available because it is too inconvenient to press any buttons on the phone during the game Players have to mount a phone holder on their bike and put something under the front wheel to turn the wheel more comfortably The handlebar turn control does not feel natural, as in reality, steering only starts by turning the handlebars and continues by tilting the body, and this handlebar turn is only done by a few degrees Only the turn movement is available because it is too inconvenient to press any buttons on the phone during the game Although new controllers like Zwift Play offer some improvements, they are product-specific and don’t solve the issue universally. Game Availability and Reliability Maintenance announcements in video games are not uncommon, but they are unacceptable in training software. Having to reschedule your workout due to maintenance can be highly annoying. Imagine an athlete who has already put on their cycling kit, filled a water bottle, and set their mind on a training session. Needless to say, the churn rate spikes dramatically whenever the system is unavailable, and it’s even higher if there’s a crash during a workout. Hardware Communication Many companies produce smart trainers and other equipment for indoor cycling, and it’s not uncommon for users to combine different brands in their setup. The problem is that companies sometimes utilize various communication protocols. Even when they claim to follow standard protocols, their implementations may vary. When users may own so many possible combinations of sports equipment, it isn’t easy to ensure reliable communication and seamless compatibility. The Next Big Move After spending some time on the projects in this area, it became evident that additional specialized equipment is required for a truly engaging indoor cycling experience. Ideally, it should be a third-party solution compatible with different equipment, providing different interaction methods. It turned out that such a platform already exists—it’s a VR headset that is almost a perfect fit for indoor cycling. It’s always astonishing when an emerging technology matures so that new applications adopt it. A VR headset not only enables the use of natural movements as input while cycling, but it also comes equipped with all the necessary hardware features for an optimal indoor cycling platform, including: An all-in-one hardware unit, eliminating the need for a separate TV or PC setup High-quality audio input and output Engaging visual capabilities Robust Bluetooth connectivity for external hardware support An all-in-one hardware unit, eliminating the need for a separate TV or PC setup An all-in-one hardware unit, eliminating the need for a separate TV or PC setup High-quality audio input and output High-quality audio input and output Engaging visual capabilities Engaging visual capabilities Robust Bluetooth connectivity for external hardware support Robust Bluetooth connectivity for external hardware support Of course, some challenges exist, such as reduced ventilation, sweating, and limited battery life. However, based on my experience, these issues are manageable and don’t affect my enthusiasm for the notion that the future of gamified indoor cycling lies in VR. I’d love to hear your thoughts on this! I’d love to hear your thoughts on this!

Stories, News & Mentions about ATAR

Atar’s Stories on HackerNoon