A model-free RL algorithm that ensures stable and efficient policy updates by optimizing within a trust region, limiting the step size to prevent drastic policy changes and improve convergence. ...

Notes on DQN and its variants. ...

Natural Evolution Strategies, or NES, are referred to a family of evolution strategies that throughout its generations update a search distribution repeatedly using an estimated gradient of its distribution parameters. ...

Notes on Policy gradient methods. ...

Covariance Matrix Adaptation Evolution Strategy (CMA-ES) is an evolutionary algorithm for complex non-linear non-convex blackbox optimization problems in continuous domain. ...

(WIP) Note III of the measure theory series. Materials are mostly taken from Tao’s book, except for some needed notations extracted from Stein’s book. ...

Notes on using linear models in regression and classification. ...