Kevin On
I’m an MS student at Stanford, mostly interested in robotics. Also into AI products, chess, and soccer.
Chess.com:Posts
Reinforcement Learning series
03. Q-learningJul 12, 2026
From Bellman optimality and DQN to DDPG, TD3, maximum-entropy RL, and SAC.
Reinforcement Learning series
02. Policy Gradient MethodsJul 10, 2026
From REINFORCE to actor-critic, GAE, TRPO, and PPO.
Reinforcement Learning series
01. Foundations of Reinforcement LearningJul 9, 2026
From MDPs and policies to value functions and Bellman equations.
Beyond Human Priors in RoboticsJun 1, 2026
Thinking through simulation, touch, and human priors in robotics.
codfish #1: Building a Chess EngineApr 15, 2026
Starting a single-GPU chess engine training project
My Experience with "OS in 1000 Lines"Jan 4, 2026
A hands-on review of the OS in 1000 Lines tutorial
Starting a BlogDec 21, 2025