Elusiv

elusiv

Authors:
(1) Antonio Riotto, Département de Physique Theorique, Universite de Geneve, 24 quai Ansermet, CH-1211 Geneve 4, Switzerland and Gravitational Wave Science Center (GWSC), Universite de Geneve, CH-1211 Geneva, Switzerland;
(2) Joe Silk, Institut d’Astrophysique, UMR 7095 CNRS, Sorbonne Universite, 98bis Bd Arago, 75014 Paris, France, Department of Physics and Astronomy, The Johns Hopkins University, Baltimore MD 21218, USA, and Beecroft Institute of Particle Astrophysics and Cosmology, Department of Physics, University of Oxford, Oxford OX1 3RH, UK. Table of Links Abstract and 1 Introduction 2 Some open questions 2.1 What is the abundance of PBHs? 2.2 What is the effect of PBH clustering? 2.3 What fraction of the currently observed GW events can be ascribed to PBHs? 2.4 Are PBHs the Dark Matter? 3 The PBH Roadmap 3.1 High redshift mergers 3.2 Sub-solar PBHs 3.3 Plugging the pair instability gap with PBH? 3.4 PBH eccentricity, 3.5 PBH spin and 3.6 Future gamma-ray telescopes 4 Conclusions and References 2.1 What is the abundance of PBHs? The formation of a PBH in the early universe is a rare event and knowing the precise formation probability represents, even nowadays, a challenge. This issue is of fundamental importance as the formation probability is one of the key ingredients to calculate the current PBH abundance and mass function which not only enter in many observables, such as the merger rate of BH binaries, but which are also routinely used to express current constraints from various observations. To illustrate the difficulty in assessing the precise abundance of PBHs, let us consider the most standard scenario in the literature where the PBHs are formed by the collapse of large inhomogeneities generated during the inflationary stage. The latter delivers a stochastic quantity, the curvature perturbation ζ on super horizon scales, whose properties are known statistically. As we mentioned, PBHs are rare events which require a given observable, e.g. the density contrast δ or the compaction function C, to be above a typically large critical value. This implies that the calculation of the PBH abundance requires going beyond linear perturbation theory and the knowledge of the tail of the probability where the latter is typically far from being a simple Gaussian distribution. Nonlinearities enter into the game even when the initial curvature perturbation is Gaussian (which is typically not) because of the nonlinear relation between the over density/compaction function an the curvature perturbation itself. The logic chain is (in radiation) but no calculation exists in the literature so far and no understanding of why the volume average of the compaction function has a critical threshold which does not depend on the profile. Even without the need of mentioning other possible problems, e.g. dealing with the nonlinearities entering in the radiation transfer function when perturbations re-enter the horizon [3] or the need to deal with the phenomenon of operator mixing [4], we believe the calculation of the PBH abundance in the standard scenario is still an open issue which deserves further study. The same is true in alternative mechanisms to generate the PBHs, e.g. in supercooled first-order phase transitions [5] where the threshold criterion does not suffice. This paper is available on arxiv under CC BY 4.0 DEED license. Authors: (1) Antonio Riotto, Département de Physique Theorique, Universite de Geneve, 24 quai Ansermet, CH-1211 Geneve 4, Switzerland and Gravitational Wave Science Center (GWSC), Universite de Geneve, CH-1211 Geneva, Switzerland; (2) Joe Silk, Institut d’Astrophysique, UMR 7095 CNRS, Sorbonne Universite, 98bis Bd Arago, 75014 Paris, France, Department of Physics and Astronomy, The Johns Hopkins University, Baltimore MD 21218, USA, and Beecroft Institute of Particle Astrophysics and Cosmology, Department of Physics, University of Oxford, Oxford OX1 3RH, UK. Authors: Authors: (1) Antonio Riotto, Département de Physique Theorique, Universite de Geneve, 24 quai Ansermet, CH-1211 Geneve 4, Switzerland and Gravitational Wave Science Center (GWSC), Universite de Geneve, CH-1211 Geneva, Switzerland; (2) Joe Silk, Institut d’Astrophysique, UMR 7095 CNRS, Sorbonne Universite, 98bis Bd Arago, 75014 Paris, France, Department of Physics and Astronomy, The Johns Hopkins University, Baltimore MD 21218, USA, and Beecroft Institute of Particle Astrophysics and Cosmology, Department of Physics, University of Oxford, Oxford OX1 3RH, UK. Table of Links Abstract and 1 Introduction Abstract and 1 Introduction 2 Some open questions 2 Some open questions 2.1 What is the abundance of PBHs? 2.1 What is the abundance of PBHs? 2.2 What is the effect of PBH clustering? 2.2 What is the effect of PBH clustering? 2.3 What fraction of the currently observed GW events can be ascribed to PBHs? 2.3 What fraction of the currently observed GW events can be ascribed to PBHs? 2.4 Are PBHs the Dark Matter? 2.4 Are PBHs the Dark Matter? 3 The PBH Roadmap 3 The PBH Roadmap 3.1 High redshift mergers 3.1 High redshift mergers 3.2 Sub-solar PBHs 3.2 Sub-solar PBHs 3.3 Plugging the pair instability gap with PBH? 3.3 Plugging the pair instability gap with PBH? 3.4 PBH eccentricity, 3.5 PBH spin and 3.6 Future gamma-ray telescopes 3.4 PBH eccentricity, 3.5 PBH spin and 3.6 Future gamma-ray telescopes 4 Conclusions and References 4 Conclusions and References 2.1 What is the abundance of PBHs? The formation of a PBH in the early universe is a rare event and knowing the precise formation probability represents, even nowadays, a challenge. This issue is of fundamental importance as the formation probability is one of the key ingredients to calculate the current PBH abundance and mass function which not only enter in many observables, such as the merger rate of BH binaries, but which are also routinely used to express current constraints from various observations. To illustrate the difficulty in assessing the precise abundance of PBHs, let us consider the most standard scenario in the literature where the PBHs are formed by the collapse of large inhomogeneities generated during the inflationary stage. The latter delivers a stochastic quantity, the curvature perturbation ζ on super horizon scales, whose properties are known statistically. As we mentioned, PBHs are rare events which require a given observable, e.g. the density contrast δ or the compaction function C, to be above a typically large critical value. This implies that the calculation of the PBH abundance requires going beyond linear perturbation theory and the knowledge of the tail of the probability where the latter is typically far from being a simple Gaussian distribution. Nonlinearities enter into the game even when the initial curvature perturbation is Gaussian (which is typically not) because of the nonlinear relation between the over density/compaction function an the curvature perturbation itself. The logic chain is (in radiation) but no calculation exists in the literature so far and no understanding of why the volume average of the compaction function has a critical threshold which does not depend on the profile. Even without the need of mentioning other possible problems, e.g. dealing with the nonlinearities entering in the radiation transfer function when perturbations re-enter the horizon [3] or the need to deal with the phenomenon of operator mixing [4], we believe the calculation of the PBH abundance in the standard scenario is still an open issue which deserves further study. The same is true in alternative mechanisms to generate the PBHs, e.g. in supercooled first-order phase transitions [5] where the threshold criterion does not suffice. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv

The Elusive Abundance of Primordial Black Holes

What is Zero-knowledge proof? As per , Zero-knowledge proofs are used to establish the truth of a claim without disclosing the claim itself. The “prover” is the person making an assertion, while the “verifier” is the party checking that assertion. In a 1985 work by Shafi et al, titled “ ,” the term “zero-knowledge proof” was first used along with the following definition. The goal of a zero-knowledge protocol is to allow one party (the prover) to demonstrate the truth of a statement to another (the verifier) without disclosing any information beyond the fact that the statement in question is true. The development of zero-knowledge proofs over the years has resulted in their widespread usage in a variety of practical contexts. Ethereum Foundation The knowledge complexity of interactive proof systems The Necessity For Zero-Knowledge Proofs Begs an Obvious Question: Why? As per Ethereum Foundation, a major step in the field of applied cryptography, zero-knowledge proofs promise to make personal data safer. Think about the steps you would take to convince someone of anything you assert (such as, “I am a member of Westminster nation”). A national passport or driver’s licence would suffice as “proof” to support your claim. forward However, this method has certain drawbacks, the most significant being a loss of privacy. Shared personally identifiable information (PII) is in centralised databases that hackers may easily access. There have been requests for more private ways to share information in light of the growing problem of identity theft. maintained This issue is addressed by zero-knowledge proofs, which do not need secret information to establish the veracity of a claim. The statement (a “witness”) is sent through the zero-knowledge protocol, which then outputs a brief justification for the statement’s veracity. The information needed to create the evidence is kept secret while providing strong assurances that the assertion being proven is accurate. Referring to our original illustration, a zero-knowledge proof is required to back up a claim of citizenship. The verifier may be persuaded that the underlying assertion is true by only confirming that specific aspects of the evidence are actual. Elusiv raises $3,5 million in a seed funding round co-led by LongHash Ventures to build Compliant Privacy Protocol , a privacy protocol that is Zero-Knowledge based and compliant, has raised $3.5 million in a seed investment round co-led by and Staking Facilities Ventures. It plans to utilise the money to hire more programmers, pay for security checks, and adapt its technology to work with many other blockchains. Elusiv LongHash Ventures Several angels, such as the founders of Solana, Zeta, Notify, Solflare, and UXD Protocol, have also invested. Other investors include Jump Crypto, NGC Ventures, Big Brain Holdings, Anagram, Cogitent Ventures, Equilibrium, Marin Ventures, Token Ventures, Moonrock Capital, Monke Ventures, and SolanaFM. : Vested Interest Disclosure The author is an independent contributor publishing via our . Be it through direct compensation, media partnerships, or networking. The author has a vested interest in the company/ies mentioned in this story. HackerNoon has reviewed the report for quality, but the claims herein belong to the author. #DYOR brand-as-author program What is Elusiv? Compliance is at the heart of Elusiv, a Zero-Knowledge privacy system for anonymous on-chain financial transactions. Elusiv protects its users by preventing hostile actors from accessing the privacy protocol while still granting them privacy (even while engaging with regular non-private wallets on the blockchain). On top of that, Elusiv VMs, a new addition to Elusiv, allow for the quick creation of solutions that use broader ZK circuits. Don’t forget to like and share the story! Image credits: and . Austin Distel Mo

Elusiv Raises $3.5 million in Seed Funding Round Co-led by LongHash Ventures to Build Web3 Privacy

Every large company brags about their superior grasp of the concepts of work-life balance and their abilities to achieve it for their employees. In my experience, very few actually achieve it. To be fair, though, it’s not really the company’s job to achieve a work-life balance. It’s their job to support the concept, it’s on us as employees to achieve it. Missing Out My five-year-old son and I had this conversation one Friday night while he was eating dinner, and I was preparing to head into my office for a late call. Son: “Are you all done with meetings today?” Me: “No, I have one left.” Son: “I’m growing up, and you’re missing it. You should call your manager and tell him you quit.” is a little bit complex for a child of his age, so I’m not exactly sure where he heard this, but I’m assuming he’s quoting some movie he saw. However, the truths in his statement still stung. The harsh truth; I’m missing out by spending too much time at work is hard to digest. “I’m growing up and you’re missing it” Let’s shed some additional light on this situation before we overdramatize it, though. First, I work from home and have for several years. I try to be very diligent about “work” time vs “home” time, but I don’t have a commute and come out to spend lunchtime with the family, so I probably get to spend more time with him than most fathers. Second, my wife is a stay-at-home-mom, so she’s here with him all day. Due to the pandemic, she is homeschooling him, so they spend a fair amount of time together. I really have to assume that this comment, if it isn’t him just repeating a TV show, comes from the sheer fact that mom spends more time with him than I do. That being said, I’ve always battled with the concept of work-life balance. Finding Balance Early in my career, and for the first year or two at my current company, I worked a lot. I received projects with tight deadlines, sometimes ridiculous deadlines, and I met each and every one.  My manager created mini competitions to see who could reach an unrealistic deadline first, and I fed into that like a starved lion. I would check my email first thing in the morning. I would even run some tests or write a little code before breakfast. I’d commute the forty-five minutes to work and complete a full 9-5 workday. I’d then commute home, have dinner with the family, then when everyone was in bed, I would work some more. My manager needed daily, sometimes hourly updates on our progress.  He wasn’t a developer by trade and rarely understood what we were talking about,  but he insisted on hearing all of the details anyway. Some days it felt like he wanted a printout of every line of code I looked at for the day. But I trudged on.  I met my deadlines, sometimes I crushed them, but I saw the red haze of burnout looming on the horizon, and I could tell my time at this company was short if I didn’t make a change. One summer, within those first two years at this company, my manager took a two-week vacation to compete in a bike race. During those two weeks, I took full advantage of not having someone breathe down my neck at all hours. I worked 9-5 and nothing else. I caught up on sleep. I resumed some of my hobbies. I had more energy to enjoy time with my family. And something else amazing happened. I not only hit my tight deadlines, but I completed an additional two projects within the same deadline. When my manager returned, he was amazed at what I had accomplished and bragged in many meetings about the three projects I had completed in a record amount of time. And for a while, I couldn’t quite place my finger on what had happened, other than the fact that I hit a great stride in my workday because I was free from his constant interruptions.  But what really happened? I had relaxed. I momentarily stopped walking into every day carrying the stress of the weeks prior and the weeks ahead on my shoulders. I woke up and viewed every day with fresh eyes. I committed time to things outside of work and actually found myself excited to be there when I finally got to my desk. I cleared my mind of all the rubble that was blocking my rational thought and allowed myself to really focus on the problems at hand… and when I did, they weren’t nearly as complicated as they seemed. I had found a brief oasis of work-life balance. Demands Of course, my manager returned, and with him came all the stress that a normal micromanager brings to the table.  Now, this specific situation is dramatic, micromanagers aren’t the only source of work-life balance issues, but it was in this specific situation. There are others. A couple of years went by, and I found myself on a new team with a new manager. This manager was not the micromanager I had before, but he was incredibly demanding, as was his manager and the manager above him. Their entire chain was wrought with unrealistic demands, and everyone was scrambling to make their superiors happy and I had been thrown right in the middle of it.  I returned to working long hours, nights, weekends, I even pulled an all-nighter to bring a production system up on a new stack for no real reason other than I was told to. (Yes told, not asked). And at every step of the way, the rest of the team and I were constantly told (in public settings) that we weren’t good enough. That we had to work harder. That our work was not acceptable and not up to the company standard. We were pushed harder and harder, and no matter what you accomplished, it could have always been better. We were publicly shamed, humiliated, forced to demo broken features or bugs. And all the while, we were asked to put in extra hours assuming if we did it would eventually get better. It did not. When it comes down to it, our customers were happy.  They loved the tools we were creating. The applications brightened their days and made their lives easier, but to our management chain they were a blight on the company’s good record. And so we worked. The Problem As I stated at the top of this stream of consciousness that I’m attempting to call an article, the company can support work-life balance, but it’s not responsible for maintaining it. That falls on the individual employees, and here is where most of us get stuck. My managers were employees. Their managers were employees. Employees who did not have a work-life balance, and so did not believe their reports deserved it either. So they pushed their employees to work. My first manager, the micromanager I discussed, also had work-life issues of his own to consider. He would often talk about how he had so many meetings in a day he didn’t have time to answer emails until 11 PM. He reserved 11-12 PM as email time, and it was not uncommon to receive vital feedback you were waiting on at that time (and he expected you to respond and take action on it at that time as well, but that’s for another article). I believe this issue boils simply down to a construct very similar to jealousy.  As managers adopt more and more responsibility, they see their ability to control work-life balance slip away, and so they jealously steal it from those below them. Force everyone to work as hard as they do so that they don’t feel like they’re failing themselves or their families. “This is just how it is here” or “we work hard we play hard” are common sayings amongst these people. But let’s be clear, all this is is a failure to properly manage your time.  The higher up you go in a company, the more pressure you have to produce results, and thus the more pressure you put on those below you to work. The Solution? I add a question mark there because in my 12 years of experience, I can’t say for sure that I’ve figured it out. The brief periods where I’ve felt that I had a good balance, I always worried that that balance would make me look lazy and cause me to lose my job. This is due to the pressure on my managers to produce unrealistic results, yes, but it’s still a concern.  Further, when I did achieve a great balance that I discussed earlier while my manager was on vacation, I tried to retain that balance upon his return. Of course, this was met with failure and even started to jeopardize my career slightly as my manager saw an employee who used to be “all in” suddenly pulling away.  It was easy to assume this was my fault at the start. I was too quick to agree to work long hours and reach unrealistic deadlines but when you really think about it, what was the alternative? Missing deadlines and potentially being fired? No, that’s not an option at all. We do what we need to to survive, and that is what I did, so I don’t believe the blame rests on me but with those making the unrealistic demands. The point that I’m trying to make is, if your company doesn’t have a clearly defined path to success with work-life balance, anyone who achieves it will be held in a negative light, so how do we solve this? In my opinion, you have to do two things. (Note: failure to do one isn’t a complete failure, but I do think achieving both is a path to true success). 1) Be upfront at the start of your career at a new company on your work-life balance and rights. I don’t mean that you confront your manager and tell him you’ll never work a second over 8 hours a day. In fact, I don’t suggest having a conversation at all. I think you should walk into the position assuming you have a right to a work-life balance. At 4:30 PM, when someone asks you for something, tell them “Sure thing, I’ll get on it tomorrow” instead of working late. When a manager asks you for an estimate, be honest. Don’t try to impress them, don’t try to create something unrealistic to beat out the competition. Be. Real. 2) Only bend the rules when it is absolutely required This one may sound stupid, weird, or obvious at the start, but think about how many times you have done this or been asked to do this, and suddenly it doesn’t seem so obvious. Even in situations where I have been very blunt about my hours and ensured that my estimates are realistic. Even when I’ve made sure to be upfront when I don’t think a deadline is feasible, there are situations where you will find yourself compelled to put in that extra work. Either you’re excited about a project, or you want to impress someone to earn that big promotion, or you just can’t think of a good reason to say ‘no’. We have all been there. I have learned that it is best to walk into a conversation saying “lets get more information” while assuming the answer is “no I will not do this right now.” This achieves two things. First, the customer feels heard, and second, you’re not committing to anything until you have the details.  The next step is to determine emergency status. If something is a genuine emergency, then work on it. Who cares if it’s 4:30 PM on a Friday. When you took the job you agreed to keep the lights on and that’s what you should do. But if it’s not an emergency, it can wait for Monday (or at least tomorrow). So what constitutes an emergency? Emergencies are different for every application, but the general rule of thumb (for me) is: Can everyone do their job? That’s it. Super simple. Can. Everyone. Do. Their. Job? If Joe in accounting needs to take an extra 5 minutes using a workaround to complete his $5M forecast, fine. If Joe in accounting literally cannot complete his forecast and the CEO is going to fire him (and probably you), then that is an emergency. As usual, this is a completely overdramatized situation, but the fact is: assess whether or not you are actively impacting anyone’s job before committing to after-hours work.  If there is no workaround and the bug/issue/feature is critical, then it might be time to bite the bullet and just work on it. If there’s a workaround (even an annoying one), it can wait. You might believe people hate you for it, but if it were them, they'd expect the same leniency. Wrap up Work-life balance is a problem. It’s not up to your employer to solve it, but it’s not up to you either. We all need to work together to set realistic expectations to move forward happily and healthily.

The Elusive Work-Life Balance

In 2016, Google photos classified a picture of two African-Americans as “gorillas.” Two years later, Google had yet to do more than from its database of classifications. In 2016, it was shown that Amazon was to European-American consumers. In Florida, algorithms used to recommend detention and parole decisions on the basis of risk of recidivism were shown to have a , such that African-Americans were more likely to be incorrectly recommended for detention who would not go on to re-offend. When translating out of a language with gender-neutral pronouns, and into languages with gendered pronouns, Google’s word2vec neural network into translations, such that pronouns become “he” when in conjunction with “doctor” (or “boss,” “financier,” etc.) but become “she” when translated in conjunction with “nurse” (or “homemaker,” or “nanny,” etc.). remove the word “gorillas” disproportionately offering one-day shipping higher error rate among African-Americans injects gender stereotypes These issues arise from a constellation of causes. Some are underlying social roots; if you train a machine learning algorithm on data created by biased humans, you’ll get a biased algorithm. Some are simply statistical artifacts; if you train a machine learning algorithm to find the best fit for the overall population, to the extent that minorities are different in a relevant way, their classifications or recommendations will necessarily have poorer fit. And some are a combination of the two: biased humans lead to biased algorithms that make recommendations which reinforce unjustified stereotypes (for instance, harsher policing of poorer neighborhoods leads to more crime reports in those neighborhood. More crime reports trigger policing analytics to recommend deploying more cops to those neighborhoods, and voila! You have a nasty feedback loop). The trouble is that it’s not at all clear how to make algorithms fair. And in this regard, conversations about algorithmic fairness have been a magnifying mirror on society’s ethics. Debates over how to define and measure fairness reflect broader ethical conversations taking place today. I’ve recently had the pleasure of interviewing , the executive director of the at Stanford. We got to talk about some of his applied work in algorithmic fairness. In particular, we got to discuss the perks and shortcomings of three sides of the debate over how to conceptualize fairness algorithmically. Technical folks can find a fuller treatment of the debate in , but I’m going to try and boil it down here. Sharad Goel Computational Policy Lab this paper Three conceptualizations of fairness This mode of thought maintains that algorithms should not be allowed to take certain protected categories into account when making predictions. For instance, in this view, algorithms which are used to predict loan qualifications or recidivism should not be allowed to base predictions off of race or gender. This approach to achieve fairness is straightforward and easy to understand. But there are two main problems: Certain group labels should be off limits. 1. . Even when such categories are eliminated from an algorithm, the statistical variance explained by these protected categories tends to slip into other available variables. For instance, while race might be excluded from loan applications, zip code, which tends to be highly correlated with race, can take on a higher predictive weight in the model and mask discrimination. For all intents and purposes, zip code becomes the new race variable. It’s challenging and debatable which proxies are illegitimate substitutes for protected categories, and which are acceptable, distinct variables. This fuzzy line brings us to the other problem with making certain labels “off-limits.” Distinguishing between acceptable and unacceptable proxies of protected categories 2. .  Protected categories often can make a meaningful impact on the behaviors that the algorithms are designed to predict. For instance, it is commonly known that insurance premiums are higher for male drivers, because male drivers really do account for more of the total insurance payouts. Eliminating gender from these algorithms would cause car insurance premiums to decrease for men, but it would increase the rates for women. Whether or not women should be required to pay for more than their share of risk, such that gender is eliminated from risk algorithms, is debatable. In short, while this may create exact equality, this seems to be missing the mark of what is proportionally equitable. Some would argue this approach is actually unfair. The societal (and sometimes personal) costs are high Higher stakes can be found in criminal justice settings. Removing protected categories like gender or race from algorithms designed to predict recidivism deteriorates the efficiency of the algorithm, meaning that more people of lower real risk are detained, and more people of higher real risk are let free. The consequences would be that more crime takes place in general, and among communities that are already experiencing higher crime in particular. To see this, keep in mind that the majority of violent crime occurs between people who already know each other. And so, communities already plagued by violent crimes may stand to experience the additional re-offending violent crimes when algorithmic efficiency is slashed (when protected, but nonetheless explanatory, categories are disallowed). Most people agree (including the law) that basing decisions on protected categories when there is no tangible justification is morally reprehensible. The tough part is when using these protected categories appears to efficiently cut down harmful outcomes. This trade off has led some to take alternative approaches to defining fairness algorithmically. Is there a way to maximize predictive accuracy (allowing inclusion of meaningful protected categories), while still being fair? As opposed to ignoring protected categories like race and gender (e.g. being color or gender blind), this approach to fairness instead argues that indicators of an algorithm’s performance should be equivalent across the protected categories. For example, an algorithm which classifies criminals as either high or low risk of re-offending should make prediction errors equally for white and black criminals. This approach is less intuitive than the color-blind approach, but at least theoretically allows the algorithms to be more efficient in their predictions, and has the added perk of avoiding tricky judgments calls about which proxies (e.g. zip code as a crude substitute for race) are and aren’t acceptable for inclusion in algorithms. Algorithmic performance should work equally well across certain groups. Still, this approach is imperfect. To see why, it’s important to understand that different groups of people will represent distinct populations—populations with different average scores, deviations, skews, kurtosis, etc. (see image above, and imagine trying to get one algorithm to perform equally for each group curve using the same cutoff threshold) . Generally, when we speak about fairness, we want all people, regardless of their group membership, to be held to the same standards. But if the same cutoff thresholds are used for different populations, predictive ability and error rates are more than likely to differ across groups--this is simply the natural result of how statistics works. If government regulation compels corporations to turn out algorithms that maintain the same performance across protected groups, corporations and institutions are incentivized to discriminate under the obscuring power of statistical wizardry and employee NDAs. They generally have two options: 1. Lower the quality and efficiency of their algorithms by toying with the code so that algorithmic performance is equal across groups (this option introduces the potential for harm discussed previously, such as releasing criminals with real, high risk scores), or 2. Corporations could adopt different algorithmic thresholds for different populations, such that cutoffs are different for different groups (genders, races, people of different sexual orientations, etc.). But clearly this seems to break with notions of fairness, and is usually morally frowned upon and considered illegal (with a notable exception being something like affirmative action). Negative impacts of forced equalization of algorithmic performance across groups aren’t just theoretical—these negative impacts have been documented, for instance, in recidivism risk score databases as well as databases predicting among white and black citizens. likelihood of police finding contraband A third approach to achieving fairness in algorithms, is to ensure that an algorithm’s scores mean equivalent things across protected categories (for instance a woman receiving a risk score of X on her insurance application, should have similar insurance payouts as a man who also receives a risk score of X on his insurance application). On the surface, it would seem that this approach is getting at what we want—it seems fair. The problem is that it cannot guarantee fairness in the presence of intentionally discriminatory action, and thus regulation of algorithms on the basis of this definition of fairness will still leave room for obscured discriminatory treatment. There are at least two ways this can happen: Algorithmic scores should represent the same things across members of different groups. 1.	Proxies (like zip code for race) can still be used to gerrymander population scores above or below an algorithm’s cutoff thresholds. For example, individuals at a higher risk of loan defaulting can be paired with individuals at a lower risk of loan defaulting, such that a protected category’s risk scores can be pushed above or below a cutoff threshold at will. This essentially boils down to algorithmic redlining. 2.	As discussed above, different groups will have different statistical risk curves. If quantitative scores are discretized (for instance, substituting “high,” “medium,” or “low” labels in place of an individual’s exact score) within groups, these differences in the real risk curves can mask different group cutoffs while maintaining the veneer that individuals labeled “high” risk re-offend, default, and get in car crashes at similar rates across protected (race, gender, etc.) categories. For example, in the image above, assigning a person a "high," "medium," or "low" risk label on the basis of their within-group percentile will effectively yield different  group cutoff thresholds, while potentially maintaining the same algorithmic performance across those labelled "high" risk for each protected group. While it seems like using these techniques would be somewhat rare for B2C corporations, who would more often than not suffer the loss of profits by discriminating in these ways, incentives still exist for B2B corporations. For instance, ad-matching companies have incentives to push certain groups above and below cutoff thresholds in order to justify ad targeting on the basis of protected categories. It’s not difficult to imagine political campaigns or lobbyists being attracted to the power of these methods to sway public opinion among strategic subgroups while leaving behind few breadcrumbs, and convoluted breadcrumbs at that. (I'm just saying, if US , my faith in their understanding this issue is...well it's not good.) senators couldn't understand Facebook's business model The challenge Each approach to algorithmically defining fairness has its strengths and weaknesses. I think what’s most troubling is not so much the weaknesses that each approach faces, but instead that these approaches are fundamentally incompatible with one another. We cannot ignore protected categories while using protected categories as the baseline to detect fairness. And we can’t demand similar algorithmic error rates while demanding that similar risk scores actually do entail similar outcomes among groups. The race is still on to define fairness algorithmically. But my background in moral psychology also gives me pause. Democrats, Republicans, and Libertarians can't agree on what's fair, and I think it's too optimistic to treat algorithmic fairness like a mathematical, computer science problem. The trouble isn’t solving some complicated statistical rubix cube, so much as it is trying to manifest Plato’s perfect form of fairness on a cave wall that’s only capable of capturing shadows. It’s hard to predict which solutions we’ll embrace, and what the costs will be when those solutions interact with regulatory and economic incentives. Algorithmic fairness is, at its heart, a socio-moral problem. I'm building out the egalitarian infrastructure of the dWeb with ERA . If you enjoyed this article, I'd love to connect with you on Twitter !

Why Algorithmic Fairness is Elusive

Despite the popular notion, Blockchains are not that complex once you get the general idea behind them. What makes Blockchain so interestingly difficult to implement is that it is made up of some many different components of varying complexity and a good Blockchain developer needs to know how to connect these components in the right order to build good Blockchains. In this post, I will explain what a Blockchain actually is, then start talking about various features of Blockchain such as Hashing, Immutability, and Distributed P2P Network. This is Part 1 of 2 of my series on Blockchain. Part 2 will be out soon! Defining Blockchain The credit for inventing Blockchain technology is generally given to an anonymous entity (could be one person or a group, nobody knows) named Satoshi Nakamoto and his academic paper titled published in 2008. “Bitcoin: A Peer-to-Peer Electronic Cash System” But there were two other people before Satoshi who spoke about something very similar to Blockchain. These individuals were Stuart Haber and Scott Stornetta and they released a paper titled “ ”. How to timestamp a digital document While this paper doesn’t openly discuss anything related to Blockchain, several concepts, features, and ideas of what makes Blockchain… Blockchain are present in this paper. But what is a Blockchain? If you go to Wikipedia — the universal collection of definition, you will see that it is defined as “ ”. a growing list of records, called blocks, which are linked using cryptography In simple terms, it is a list of blocks, where the blocks are linked to each other in the form of a chain. The next question in your mind should be: What is a Block? A Block in a Blockchain is defined as a record that consists of some data, a pointer called that addresses the block that comes before it, and another pointer called that addresses the block itself. previous hash hash Here is an example of a simple blockchain made of 2 blocks: The first block is called the Once this block is created, it will stay the same for all of eternity. There is no way, any other block can take its place. Like any other block, it will have some data stored in it. But because it is the first block in the chain, and there is obviously no block before it, its previous hash value will be a string of zeros. The block also has a hash value, which is a string of hexadecimal characters. genesis block. The second block will also have some data stored inside it. And it’s previous hash value will be the same as the previous block’s hash value. Apart from that, it will also have a hash value which will the previous hash value of the next block in the chain, and so on. The hashes are what link the blocks to each other. Each block knows its own identity and the identity of the block before it. This method of linking also secures the blockchain from any changes to the data of the block. The hash of a block is created by taking the data stored inside it and the previous hash. So if I was somehow able to change the data of one block, then that block’s hash will also change. But now the next block’s previous hash value will not match with the new hash value of the changed block. And then the blockchain will know that the block has been corrupted, and the blockchain will then become . invalid Hashing The previous few paragraphs describe how hashing connects the block to each other, and how it can also be used to secure the blocks. But what exactly is Hashing? When it comes to human beings, there are a couple of things that can be used to differentiate and identify each of them: DNA Fingerprint And while identical twins share the same DNA, they do not have the same fingerprints. So fingerprints can be considered as the ultimate unique identifier of any individual. Hashing is a blockchain’s version of fingerprints! There are different algorithms that can be used for Hashing, the most popular one is the SHA 256 and it is what is used in most blockchains and cryptocurrencies. SHA stands for Secure Hashing Algorithm and it has many different versions. SHA-2 is a set of cryptographic hash functions that are designed by the National Security Agency (NSA) and contains six hash functions. The “256” part of SHA-256 is used because that’s the number of bits it takes up in memory. The Hash created by this algorithm will be 64 characters long and it will be in hexadecimal format. Hexadecimal stands for 16, so each character of the hash takes up 4 bits (4² = 16, and 16*4 = 64). Hashing can be used on any kind of data, be it money, documents, programs or videos. _SHA-256 produces a 256-bit (32-byte) hash value. The SHA (Secure Hash Algorithm) is one of a number of cryptographic…_www.xorbin.com SHA-256 hash calculator. Online SHA-256 hash generator. Mining Bitcoin The above link takes you to an online hash generating website. Try entering different words in the input area and see what SHA-256 Hash you get. Here are couple that I got: Rajat -> 126e11da02566f50f0f9449586cd9069ba908ce82ebfd69d6c3b9f48f56853a5 rajat -> a2acfac793fcfd3ced6506d76fabfcaea56fcb07d59a173663f003b85465a18f Block ->211d0bb8cf4f5b5202c2a9b7996e483898644aa24714b1e10edd80a54ba4b560 You will notice that both “Rajat” and “rajat”, while being the same word, have completely different hashes due to one of them being capitalized. This complete change of hashes due to slight changes in data is known as the avalanche effect. It doesn’t matter if your use hash on a single letter, a single word, or an entire Harry Potter book, the length of the hash will always be 64 characters. There are a few things you need to keep in mind while working with a hash function: The hash function works one way only. You will not be able to figure out what the original data was using the hash. The hash function should give the same output regardless of time. So if I use hash function on “Rajat” and get some hash, and then a couple days later I do the same thing again, I should get the same hash. Hash function should give the hash quickly. Even the smallest changes in data should completely change the hash. Because the hash will always be 64 bits long, we have a limited number of potential hashes to work with. So chances are that sooner or later, you will get two blocks with the same hash value. But such collision will rarely occur and there are ways to work around them in case they happen. But hackers can also create collisions cases, which leads to further problems in blockchain. You can read more about Secure Hash Algorithms . here Immutability Immutability is one of the most popular features of Blockchain. To be immutable means to remain unchanged. So in terms of Blockchain, once you create a block, no one change make any changes to it. No even the person who created. This is a great way of maintaining trust in Blockchain. In the real world, when you buy something big, like a house, you are actually exchanging a fixed amount of money for a deed that says that the house is yours. But then anyone can write up a deed saying that the house that you just bought, is actually theirs! To prevent such practices, you go to a central authority such as the city council and register your ownership. And now you legally own that house! Similarly, in the blockchain, there needs to be some way to prevent anyone from taking unauthorized ownership of a block or the entire chain. But Blockchain doesn’t have any central authority where users can register their blocks! While you may think of this as a drawback of the blockchain, it is actually an advantage! If blockchain had a central authority that recorded any data about the blocks, then a hacker simply needs to hack into it and can then make any unauthorized changes. Instead, a blockchain is decentralized. This decentralization of systems was one of the main things that people like Satoshi Nakamoto and Vitalik Buterin (the guy behind Ethereum) spoke about. Previously, I spoke about how changing the data on a block will change its hash value. If someone manages to do this, they will have to change the previous hash value of the next block as well, which will then change the hash value of that block as well. The hacker will then have to change the previous hash value of the next hash value, and so on. So in order to change the data of one block, the hacker will have to make changes to the rest of the blocks as well! This will require some serious computing power and hence is considered to be an impossible task for an average hacker. And this how Hashing is used to implement immutability in a Blockchain! Distributed P2P Network A Blockchain network is known as a will have multiple computers (aka nodes) interconnected to each other, with no central authority. Here is an example: distributed Peer to Peer Network Each node will have a copy of the Blockchain. Here you see a small network of 5 computers. But in reality, Blockchain networks can have anywhere from hundreds to thousands of computers connected to each other. And these computers don’t have to be high-end. Any average laptop or desktop can become a node in a blockchain network. While each node has a copy of the blockchain, they cannot actually someone else’s data without proper authorization. When a node adds a block to the end of the blockchain, other nodes are notified by the network about it, and all the copies of the blockchain are updated accordingly. It’s not just hackers that we need to worry about. Sometimes, glitches occur that can lead to different kinds of error in one of the blockchain copies. Due to Blockchain’s immutability, users cannot make changes to the blocks once they are added to the blockchain. How can we then correct the chain if one of its copies shows different data than the rest, without there being any malicious force behind it? The Blockchain network is constantly syncing all the nodes with each other. So when one of the node’s copies of Blockchain gets corrupted, the network will realize that something is wrong with this copy, and the network will replace the corrupted blocks of that chain with the uncorrupted blocks from the other copies. The only way to fool the Blockchain network is by making changes to majority nodes of the network. So if a network has 100 nodes, the hacker will have to change more than 50 nodes, which is still a huge number. Hence the security of a Blockchain network increases as its number of users increases. To be continued… This seems like a good place to pause. Stick around for more posts about Blockchain. In the next post, I will explain about: Mining Concensus Thanks for reading this long post! I hope this post helped you understand Blockchains a little better. If you liked this post, then please do give me a few 👏 and please feel free to comment below. Cheers! This concludes Part 1 of 2 of my series on Blockchain. Part 2 will be out soon!

Understanding The Elusive Blockchain Technology — Part 1

Episode 11: An interview with creative entrepreneur, best selling author, and podcaster . David Kadavy Listen to the interview on , or , or watch on . iTunes Google Podcast YouTube In this episode and discuss elusive black swam moments, which are moments that describe an event that comes as a surprise and have a major effect. Trent Lapinski David Kadavy “I had an event where my e-mail list went from 5,000 to 30,000 within a couple weeks. All these events, they kind of happened, and I did things to make them happen, but I couldn’t have predicted that they were going to happen. So I wondered to myself, how can you make black swans happen?” “They suffer from interacting with an issue I call irrational rationality. Which is the treating of absence of evidence as evidence of absence.” “It always really turned me off, the whole A/B testing obsession that there was in Silicon Valley… Well yeah, if you can use machine learning to do that, that’s even more reasons for you to try and ignore those activities. Then look for the activities that are low investment, but uncertain in outcome. That uncertain outcome may be a very positive outcome.” — David Kadavy _Episode 11: An interview with creative entrepreneur, best selling author, and podcaster David Kadavy. In this episode…_www.podbean.com The Elusive Black Swan with David Kadavy P.S. If you dig the new Hacker Noon Podcast , consider giving us a 5 star review on iTunes. Also check out , and . November’s top stories the latest stories today’s homepage

The Elusive Black Swan with David Kadavy

I have taken decisions in my life where I’ve turned down job offers that offered more but required me to work longer hours. The reasoning was that I would have more free time to pursue other things that I’m interested in, like writing a blog post every day or reading a new book every week or writing jokes and performing standup comedy. money I know a lot of people who have made similar decisions — to value free time more than money (after a certain point). There have been several studies that have concluded this as well — that earning more money than a certain amount ($75000 a year in the US) has little to no effect on the happiness of people. After that point, it is the things they do with their time that can significantly impact their levels of happiness. Now, let us say, every person optimizes their lives this way. Find good work that pays well until they reach that threshold (equivalent of $75000 in the US), and then spend the rest of their time doing things they enjoy more. When people would rather spend two hours a day with their kids and reading books than earn another $25000 a year by working those two extra hours a day, you would expect the world to be a happier place. But, that $25000 that people forego will not be spent on consuming things they don’t really need. This drop in consumption results in faltering growth for companies that make these things that these people don’t really need as their profits begin to slump. As a result, the overall market cap of the economy and hence the GDP, begins to fall. This results in some of these companies cutting costs and laying off employees and/or forcing down their wages. And an equilibrium is reached only when those people that decided to spend two hours with their kids and reading books instead of working those two extra hours, now have to work those two extra hours just to make the original threshold ($75000 equivalent). And the cycle repeats. Our economy is not structured to allow people to work a few hours and have a few more hours free. Our incessant need for GDP growth results in a society where people need to constantly maximize their and fill up as much of their time with work as possible. productivity It doesn’t really matter if we don’t value the things we can buy with the extra money we can make by working extra hours or not. We need to work those extra hours just to keep the semblance of our society and economy going. There will always be a handful minority that exemplifies saying no to extra work and extra money in favour of extra free time and a significant majority looking up to them and hoping to be like them. But they cannot do so. Because, once they do, they just shift the equilibrium to a new place which again results in a handful minority and a significant majority on opposite sides as before. For the majority, free time is permanently elusive. Unless, we find a way to restructure our economy to make that graspable. Before you go… If you liked this, support my work. All you need to do is clap. Discuss the intricacies of the good life with me. Read my book

Free time is permanently elusive

From OpenAI to Closed AI: Custom Chips Are Closing The Doors—What’s Next? 

Ад OpenAI да закрытага AI: Карыстальніцкія мікрасхемы зачыняюць дзверы - што далей? 

De l'OpenAI a la IA tancada: els xips personalitzats estan tancant les portes: què passa? 

Von OpenAI zu Closed AI: Kundenspezifische Chips schließen die Türen – was kommt als Nächstes?

Από το OpenAI στο Closed AI: Τα προσαρμοσμένα τσιπ κλείνουν τις πόρτες—Τι ακολουθεί; 

De la IA abierta a la IA cerrada: los chips personalizados están cerrando las puertas. ¿Qué sigue?

Az OpenAI-tól a zárt mesterséges intelligenciaig: Az egyedi chipek bezárják az ajtókat – mi a következő lépés? 

Dari OpenAI ke AI Tertutup: Chip Kustom Menutup Pintu—Apa Selanjutnya?

OpenAI からクローズド AI へ: カスタム チップが扉を閉ざす — 次は何が起こるのか?

Kobanda na OpenAI tii na AI oyo ekangami: Ba puces personnalisées ezali kokanga ba portes —What’s Next? 

له OpenAI څخه تړل شوي AI ته: دودیز چپس دروازې بندوي — بیا څه دي؟ 

От открытого ИИ к закрытому ИИ: кастомные чипы закрывают двери — что дальше?

Від відкритого штучного інтелекту до закритого штучного інтелекту: спеціальні мікросхеми закривають двері — що далі? 

How Zen Principles Can Revolutionize Your Data Strategy

Cypherpunks Write Code: Len Sassaman & Remailers

সাইফারপাঙ্কস কোড লিখুন: লেন সাসামান এবং রিমেলার

Los cypherpunks escriben código: Len Sassaman y los remailers

Cypherpunks Kirjoita koodi: Len Sassaman & Remailers

Cypherpunks כותבים קוד: Len Sassaman & Remailers

サイファーパンクがコードを書く: レン・サッサマンとリメイラー

Cypherpunks Kokoma Code: Len Sassaman & Ba remailers

Cypherpunks Rašyti kodą: Len Sassaman & Remailers

Cypherpunks د لیکلو کوډ: لین ساسامان او ریمیلرز

Cypherpunks escrevem código: Len Sassaman e Remailers

Шифропанки пишут код: Лен Сассаман и ремейлеры

Cypherpunks Write Code: Len Sassaman සහ Remailers

ሳይፈርፓንክስ ኮድ ጽሓፍ: ሌን ሳሳማን & ሪሜይለርስ

Stories, News & Mentions about ELUSIV

Elusiv’s Stories on HackerNoon