Zac WellmerinArxiv BytesSummary: GameGANIdeas from this summary are taken from the GameGAN paper.Jun 23, 2020Jun 23, 2020
Zac WellmerinArxiv BytesSummary: Conservative Policy IterationConservative Policy Iteration has 3 goals: (1) an iterative procedure guaranteed to improve a performance metric, (2) terminate in a…May 7, 2019May 7, 2019
Zac WellmerinArxiv BytesSummary: SimPLeIdeas and figures from this summary are taken from Model-Based Reinforcement Learning for Atari(SimPLe).Apr 25, 2019Apr 25, 2019
Zac WellmerinArxiv BytesSummary: PlaNetDeep Planning Network (PlaNet), is a model-based agent that learns a latent state dynamics model from images and takes actions based on…Feb 25, 2019Feb 25, 2019
Zac WellmerinArxiv BytesSummary: World ModelsOne of the core issues in Reinforcement Learning is sample complexity. Therefore it’s appealing to train RL agents in a simulator which…Feb 16, 2019Feb 16, 2019
Zac WellmerinArxiv BytesSummary: Learning Plannable Representations with Causal InfoGANThe goal of this work is to go about planning a sequence of abstract states towards a goal and then decode the abstract states to their…Feb 1, 2019Feb 1, 2019
Zac WellmerinArxiv BytesSummary: Value Prediction Networks(VPN)VPN is a deep reinforcement learning architecture that mixes ideas from both model free and model based methods. Generally model based…Sep 15, 2018Sep 15, 2018
Zac WellmerinArxiv BytesSummary: Proximal Policy Optimization(PPO)Ideas from this summary are taken from the Proximal Policy Optimization paper.Sep 14, 2018Sep 14, 2018
Zac WellmerinArxiv BytesSummary: TreeQNIdeas from this summary are taken from the TreeQN and ATreeC paper.Sep 13, 2018Sep 13, 2018
Zac WellmerinArxiv BytesSummary: Deep Deterministic Policy GradientsThis post is a summary of Continuous Control With Deep Reinforcement Learning.Nov 10, 2017Nov 10, 2017
