Sutton & Barto (2018)

A complete walkthrough from the multi-armed bandit problem to modern deep RL and large-language-model alignment techniques (PPO, GRPO, DPO).

Standalone deep-dives

In-depth notes on individual machine learning papers and techniques, from modern optimizers to causal inference.

Osborne (2003)

Strategic interactions explored from Nash Equilibrium and mixed strategies through extensive-form games, Bayesian games, and cooperative bargaining.