Rafael StekolshchikinTowards Data ScienceEntropy in Soft Actor-Critic (Part 2)10 min read·Jun 7, 2021----
Rafael StekolshchikinTowards Data ScienceEntropy in Soft Actor-Critic (Part 1)9 min read·May 4, 2021--1--1
Rafael StekolshchikinTowards Data ScienceThree aspects of Deep RL: noise, overestimation and explorationNoise can be harmful, it can lead to a systematic overestimation. However, noise can be useful, such as noise for exploration.11 min read·May 1, 2020----
Rafael StekolshchikinTowards Data ScienceA pair of interrelated neural networks in DQNDQN and Double DQN models. In these models, comparing two interrelated neural networks is crucial.10 min read·Mar 18, 2020----
Rafael StekolshchikinTowards Data ScienceHow does the Bellman equation work in Deep RL?The connection between Bellman equation and Neural Networks, with formulas, examples and Python code11 min read·Feb 11, 2020--2--2