Affine DeFi

A comprehensive overview of advancements in text-to-image diffusion models, including personalization techniques and strides in animating these models. 

Text-to-Image Diffusion Models and Personalized Animation Techniques

Learn how algebraic geometry techniques, like schemes and rings, help identify when economic models break down or shift unexpectedly.

Detecting Instability in Economic Models Using Algebraic Geometry

Explore how the new concept of stochastic equilibrium compares with deterministic steady states and ergodic means addressing continuous time & economic dynamics

What is Stochastic Equilibrium and How Does It Change Economic Thinking?

Detailed proofs for SPD spaces, inner products, pseudo-gyrodistances, and FC layers, referencing Nguyen & Yang (2023) and Pennec et al. (2020).

Mathematical Proofs for SPD Inner Products and Pseudo-Gyrodistances in Manifold Layers

Neural networks can approximate multi-body ODEs in dynamics. Learn key concepts, normalization techniques, and challenges in deep learning-based solutions.

Neural Networks in Dynamics: Approximating Multi-Body ODEs with Deep Learning

This paper proposes a novel framework called LieBN, enabling batch normalization over Lie groups.

Proposing A Novel Framework Called LieBN: A Conclusion

Manifold-valued measurements exist in numerous applications within computer vision and machine learning.

A Lie Group Approach to Riemannian Batch Normalization

GyroSpd++ and Gr-GCN++ face limitations with hybrid methods and different Riemannian metrics, suggesting a need for optimization in SPD and Grassmann networks.

The Limitations of GyroSpd++ and Gr-GCN++ in Human Action Recognition and Graph Embedding Tasks

GyroSpd++ and Gr-GCN++ outperform baselines in human action recognition and node classification, showing superior accuracy on NTU60, FPHA, and Pubmed datasets.

New Riemannian Networks Outperform Traditional Models in Action Recognition and Node Classification

Learn how hypergyroplanes and pseudo-gyrodistance redefine fully connected and convolutional layers for SPD neural networks

Reformulating Neural Layers on SPD Manifolds

Explore how K-SIF and SIF, based on signature theory, enhance anomaly detection in complex functional data.

How K-SIF and SIF Revolutionize Anomaly Detection in Complex Datasets

A framework for privacy-preserving cloud computing, combining differential privacy and system immersion to secure data without sacrificing utility.

Privacy in Cloud Computing through Immersion-based Coding: Privacy Guarantee

Pretending to be ourselves and our friends by training an LLM on Telegram messages

I Fine-Tuned an LLM With My Telegram Chat History. Here’s What I Learned

This paper presents two neural network approaches that approximate the solutions of static and dynamic conditional optimal transport problems, respectively.

Efficient Neural Network Approaches: Partially Convex Potential Maps (PCP-Map) for Conditional OT

Dive into ImageMetalling to jazz up your images by tweaking color spaces with some cool geometric distortion tricks!

Understanding Color Space Transform Using The Moving Least Squares Method

Discover how zero-knowledge proofs (ZKPs) and Rank-1 Constraint Systems (R1CS) are transforming blockchain scalability. 

How Zero-Knowledge Proofs and Rank-1 Constraint Systems Transform Blockchain Scalability

The Geometry of UIView explained: Demystifying Frame vs. Bounds; Understanding Center and AnchorPoint; Mastering Transforms; Unraveling How Frame Is Computed

Mastering UIView Geometry: A Dive into UIKit's Foundational Class and Geometry Concepts for iOS

C++ pipeline for LiDAR-based autonomous driving.

Autonomous Driving Lidar Perception Stack with PCL: An Algorithmic Implementation

How I approached solving an interview task for autonomous driving from 3 different perspectives: RANSAC, PCA, and Ordinary Least Squares (OLS).

RANSAC, OLS, PCA: 3 Ways to Draw a Straight Line Across a Set of Points

Tree-based models like Random Forest and XGBoost have become very popular in solving tabular(structured) data problems and gained a lot of tractions in Kaggle competitions lately. It has its very deserving reasons. However, in this article, I want to introduce a different approach from fast.ai’s Tabular module leveraging.

Gain State-Of-The-Art Results on Tabular Data with Deep Learning & Embedding Layers [A How To Guide]

Late last year, my team participated in the much talked-about Y Combinator Startup School on behalf of our startup, Affine Finance. We were assigned a great mentor (former C-level executive at Zynga and a startup advisor at multiple other prominent accelerators). It was a 10-week program, with weekly live sessions conducted both in-person and online (since I was close to YC’s Mountain View offices, I ended up attending most of the sessions in-person). The live sessions touched upon a broad range of topics — everything from legal matters to design and product development.

Seven Key Takeaways from Our Journey Through Y Combinator’s Startup School

This paper describes our open source character-based language model <a href="https://github.com/IntuitionEngineeringTeam/chars2vec" target="_blank">chars2vec</a>. This model was developed with Keras library (TensorFlow backend) and now is available for Python 2.7 and 3.0+.

Chars2vec: character-based language model for handling real world texts with spelling errors and…

‘Capsule Networks’ and ‘Inverse Graphics’ seem like intimidating and somewhat vague terms when heard for the first time. These terms weren’t prevalent in mainstream media until recently, after the godfather of deep learning, Geoffrey Hinton, came out with two papers on Dynamic Routing between Capsules and on Matrix Capsules with EM Routing [This is currently a blind submission under review for ICLR 2018 but let’s be honest, we know it’s going to be Hinton et al.].

Uncovering the Intuition behind Capsule Networks and Inverse Graphics

Why would you care about some homogeneous coordinates, whatever they are? Well, if you work with geometry: 3D-graphics, image processing, physical simulation, the answer is obvious. Knowing the mathematics behind your framework of choice lets you write more efficient code. I once had to speed up one&nbsp;.NET application that bends pictures and simply re-implementing some primitive operations the right way <a href="https://leanpub.com/regainingperformanceinnetapplications" target="_blank">gave me 100 times performance boost</a>.

Programmer’s guide to homogeneous coordinates

Affine is a DeFi startup that's building a protocol for enabling cross-chain, smart yielding strategies for crypto investors. We are founded by experienced blockchain, quantitative finance, and technology experts and have raised a seed round from leading crypto investors including the founders of Dragonfly Capital and Polygon Networks.

AFFINE DEFI

affinedefi.com

Since 2021

Affine DeFi's stories on HackerNoon