Hello, I’m Jyo!

Explaining Interesting Ideas

about me

(Left) A photograph I took of a motionless jumping spider, they are very intelligent creatures. (Right) My two parakeets.

Paper TLDRs

• Discovering faster matrix multiplication algorithms with reinforcement learning
• Fractal Patterns May Unravel the Intelligence in Next-Token Prediction

Projects

• (RL) How can you guide exploration

Machine Learning

• Symmetries in Neural Networks
• Model Merging
• Sequence Modeling / Memory
• Gumbel Softmax
• Diffusion
• Transformer
• Variational Autoencoders w Implementation
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction

Statistics

• Markov chain Monte Carlo (MCMC)
  • Gibbs Sampling
  • Metropolis Hastings w Implementation
• Gaussian Process w Implementation
• KL Divergence
• D Seperation In A Bayesian Network
• Generate Uniform Points In A Circle
• Exponential distribution
• Gamma distribution
• Reservoir Sampling
• Discrete Optimal Transport
• Wasserstein Distances
• Langevin Sampling
• Thompson’s Sampling
• Importance Sampling

Reinforcement Learning

• QLearning w Implementation
• Policy Gradients
• Monte Carlo Tree Search w Implementation

Math

• Johnson–Lindenstrauss lemma TODO
• Lagrange Multiplier
• Lagrange Multiplier with Mutilple Constraints
• Why Is The Gradient The Steepest Direction? (using Lagrange Multiplier)
• Napoleon’s Theorem
• Euler Lagrange Equation

Misc

• Locality Sensitive Hashing w Implementation