Papers I Read Notes and Summaries

Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model


  • The paper presents the MuZero algorithm that performs planning with a...

Contrastive Learning of Structured World Models


  • The paper introduces Contrastively-trained Structured World Models (C-SWMs).

  • These...

Gossip based Actor-Learner Architectures for Deep RL

How to train your MAML


  • The paper proposes MAML++ - a modification of MAML algorithm that...

PHYRE - A New Benchmark for Physical Reasoning


  • The paper proposes the PHYRE (PHYsical REasoning) benchmark - consisting of...

Large Memory Layers with Product Keys


  • The paper proposes a structured key-value memory layer that:
    • Can scale...

Abductive Commonsense Reasoning


  • The paper presents the task of abductive NLP (pronounced as alpha...

Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models


  • The paper proposes a new algorithm called as Probabilistic Ensemble with...

Assessing Generalization in Deep Reinforcement Learning

  • The paper presents a benchmark and experimental protocol (environments, metrics, baselines, training/testing setup)...

Quantifying Generalization in Reinforcement Learning


  • The paper introduces a new, procedurally generated environment called as CoinRun...