Sherwin CheninTowards AIHow to Achieve Effective Exploration Without the Sacrifice of ExploitationA gentle introduction to the Never Give Up(NGU) agentJun 23, 2020Jun 23, 2020
Sherwin CheninTowards AIMuZero: Master Board and Atari Games with The Successor of AlphaZeroA gentle introductionJun 8, 2020Jun 8, 2020
Sherwin CheninTowards AIDreamer: A State-of-the-art Model-Based Reinforcement Learning AgentA brief walk-through of a state-of-the-art model-based reinforcement learning algorithmMay 31, 2020May 31, 2020
Sherwin CheninTowards AIFor The Win: An AI Agent Achieves Human-Level Performance in a 3D Video GameIntroductionJan 18, 2020Jan 18, 2020
Sherwin CheninTowards AIDNC: Differential Neural NetworkA detailed walk-through of DNCDec 8, 2019Dec 8, 2019
Sherwin CheninTowards AINeural Turing MachinesDiving Into Neural Turing MachinesDec 1, 2019Dec 1, 2019
Sherwin CheninTowards AILearn to Schedule Communication between Cooperative agentsA novel architecture for communication scheduling in multi-agent environmentsNov 16, 2019Nov 16, 2019
Sherwin CheninTowards AIScalable Efficient Deep-RLA more efficient way to scale up reinforcement learning algorithmsNov 7, 2019Nov 7, 2019
Sherwin CheninTowards AIQWeb: SolvingWeb Navigation Problems using DQNLearning to accurately navigate the webOct 11, 2019Oct 11, 2019
Sherwin CheninTowards AIModel-Based Meta Reinforcement LearningDive into a model-based meta-RL algorithm that enables fast adaptationSep 16, 2019Sep 16, 2019