Awesome Model-Based Reinforcement Learning

This is a collection of research papers for model-based reinforcement learning (mbrl). And the repository will be continuously updated to track the frontier of model-based rl.

Welcome to follow and star!

Table of Contents

• A Taxonomy of Model-Based RL Algorithms
• Papers
  • Classic Model-Based RL Papers
  • NeurIPS 2021
  • ICLR 2021
  • ICML 2021

A Taxonomy of Model-Based RL Algorithms

We’ll start this section with a disclaimer: it’s really quite hard to draw an accurate, all-encompassing taxonomy of algorithms in the Model-Based RL space, because the modularity of algorithms is not well-represented by a tree structure. So we will publish a series of related blogs to explain more Model-Based RL algorithms.

We simply divide Model-Based RL into two categories: Learn the Model and Given the Model.

• Learn the Model mainly focuses on how to build the environment model.
• Given the Model cares about how to utilize the learned model.

And we give some examples as shown in the figure above. There are links to algorithms in taxonomy.

[1] World Models: Ha and Schmidhuber, 2018
[2] I2A (Imagination-Augmented Agents): Weber et al, 2017
[3] MBMF (Model-Based RL with Model-Free Fine-Tuning): Nagabandi et al, 2017
[4] MBVE (Model-Based Value Expansion): Feinberg et al, 2018
[5] ExIt (Expert Iteration): Anthony et al, 2017
[6] AlphaZero: Silver et al, 2017
[7] POPLIN (Model-Based Policy Planning): Wang et al, 2019
[8] M2AC (Masked Model-based Actor-Critic): Pan et al, 2020

Papers

format:
- [title](paper link) [links]
  - author1, author2, and author3.
  - openreview [if the score is public]
  - key 
  - experiment environment

Classic Model-Based RL Papers

• Dyna, an integrated architecture for learning, planning, and reacting

  • Richard S. Sutton. ACM 1991
  • Key: dyna architecture
  • ExpEnv: None

• PILCO: A Model-Based and Data-Efficient Approach to Policy Search

  • Marc Peter Deisenroth, Carl Edward Rasmussen. ICML 2011
  • Key: probabilistic dynamics model
  • ExpEnv: cart-pole system, robotic unicycle

• Learning Complex Neural Network Policies with Trajectory Optimization

  • Sergey Levine, Vladlen Koltun. ICML 2014
  • Key: guided policy search
  • ExpEnv: mujoco

• Learning Continuous Control Policies by Stochastic Value Gradients

  • Nicolas Heess, Greg Wayne, David Silver, Timothy Lillicrap, Yuval Tassa, Tom Erez. NIPS 2015
  • Key: backpropagation through paths + gradient on real trajectory
  • ExpEnv: mujoco

• Value Prediction Network

  • Junhyuk Oh, Satinder Singh, Honglak Lee. NIPS 2017
  • Key: value-prediction model
  • ExpEnv: collect domain, atari

• Sample-Efficient Reinforcement Learning with Stochastic Ensemble Value Expansion

  • Jacob Buckman, Danijar Hafner, George Tucker, Eugene Brevdo, Honglak Lee. NIPS 2018
  • Key: ensemble model and Qnet + value expansion
  • ExpEnv: mujoco, roboschool

• Recurrent World Models Facilitate Policy Evolution

  • David Ha, Jürgen Schmidhuber. NIPS 2018
  • Key: vae(representation) + rnn(predictive model)
  • ExpEnv: car racing, vizdoom

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

  • Kurtland Chua, Roberto Calandra, Rowan McAllister, Sergey Levine. NIPS 2018
  • Key: probabilistic ensembles with trajectory sampling
  • ExpEnv: cartpole, mujoco

• When to Trust Your Model: Model-Based Policy Optimization

  • Michael Janner, Justin Fu, Marvin Zhang, Sergey Levine. NeurIPS 2019
  • Key: ensemble model + sac + k-branched rollout
  • ExpEnv: mujoco

• Algorithmic Framework for Model-based Deep Reinforcement Learning with Theoretical Guarantees

  • Yuping Luo, Huazhe Xu, Yuanzhi Li, Yuandong Tian, Trevor Darrell, Tengyu Ma. ICLR 2019
  • Key: Discrepancy Bounds Design + ME-TRPO with multi-step + Entropy regularization
  • ExpEnv: mujoco

• Model-Ensemble Trust-Region Policy Optimization

  • Thanard Kurutach, Ignasi Clavera, Yan Duan, Aviv Tamar, Pieter Abbeel. ICLR 2018
  • Key: ensemble model + TRPO
  • ExpEnv: mujoco

• Dream to Control: Learning Behaviors by Latent Imagination

  • Danijar Hafner, Timothy Lillicrap, Jimmy Ba, Mohammad Norouzi. ICLR 2019
  • Key: latent space imagination
  • ExpEnv: deepmind control suite, atari, deepmind lab

• Exploring Model-based Planning with Policy Networks

  • Tingwu Wang, Jimmy Ba. ICLR 2020
  • Key: model-based policy planning in action space and parameter space
  • ExpEnv: mujoco

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

  • Julian Schrittwieser, Ioannis Antonoglou, Thomas Hubert, Karen Simonyan, Laurent Sifre, Simon Schmitt, Arthur Guez, Edward Lockhart, Demis Hassabis, Thore Graepel, Timothy Lillicrap, David Silver. Nature 2020
  • Key: MCTS + value equivalence
  • ExpEnv: chess, shogi, go, atari

NeurIPS 2021

• On Effective Scheduling of Model-based Reinforcement Learning

  • Hang Lai, Jian Shen, Weinan Zhang, Yimin Huang, Xing Zhang, Ruiming Tang, Yong Yu, Zhenguo Li
  • Key: extension of mbpo + hyper-controller learning
  • OpenReview: 8, 6, 6
  • ExpEnv: mujoco, pybullet

• Model-Based Reinforcement Learning via Imagination with Derived Memory

  • Yao Mu, Yuzheng Zhuang, Bin Wang, Guangxiang Zhu, Wulong Liu, Jianyu Chen, Ping Luo, Shengbo Eben Li, Chongjie Zhang, Jianye HAO
  • Key: extension of dreamer + prediction-reliability weight
  • OpenReview: 6, 6, 6, 6
  • ExpEnv: deepmind control suite

• MobILE: Model-Based Imitation Learning From Observation Alone

  • Rahul Kidambi, Jonathan Chang, Wen Sun
  • Key: imitation learning from observations alone + mbrl
  • OpenReview: 6, 6, 6, 4
  • ExpEnv: cartpole, mujoco

• Model-Based Episodic Memory Induces Dynamic Hybrid Controls

  • Hung Le, Thommen Karimpanal George, Majid Abdolshah, Truyen Tran, Svetha Venkatesh
  • Key: model-based + episodic control
  • OpenReview: 7, 7, 6, 6
  • ExpEnv: 2D maze navigation, cartpole, mountainCar and lunarlander, atari, 3D navigation: gym-miniworld

• A Consciousness-Inspired Planning Agent for Model-Based Reinforcement Learning

  • Mingde Zhao, Zhen Liu, Sitao Luan, Shuyuan Zhang, Doina Precup, Yoshua Bengio
  • Key: mbrl + set representation
  • OpenReview: 7, 7, 7, 6
  • ExpEnv: MiniGrid-BabyAI framework

• Mastering Atari Games with Limited Data

  • Weirui Ye, Shaohuai Liu, Thanard Kurutach, Pieter Abbeel, Yang Gao
  • Key: muzero + self-supervised consistency loss
  • OpenReview: 7, 7, 7, 5
  • ExpEnv: atrai 100k, deepmind control suite

• Self-Consistent Models and Values

  • Gregory Farquhar, Kate Baumli, Zita Marinho, Angelos Filos, Matteo Hessel, Hado van Hasselt, David Silver
  • Key: new model learning way
  • OpenReview: 7, 7, 7, 6
  • ExpEnv: tabular MDP, Sokoban, atari

• MOPO: Model-based Offline Policy Optimization

  • Tianhe Yu, Garrett Thomas, Lantao Yu, Stefano Ermon, James Zou, Sergey Levine, Chelsea Finn, Tengyu Ma
  • Key: model-based + offline
  • OpenReview: None
  • ExpEnv: d4rl dataset, halfcheetah-jump and ant-angle

• RoMA: Robust Model Adaptation for Offline Model-based Optimization

  • Sihyun Yu, Sungsoo Ahn, Le Song, Jinwoo Shin
  • Key: model-based + offline
  • OpenReview: 7, 6, 6
  • ExpEnv: design-bench

• Offline Reinforcement Learning with Reverse Model-based Imagination

  • Jianhao Wang, Wenzhe Li, Haozhe Jiang, Guangxiang Zhu, Siyuan Li, Chongjie Zhang
  • Key: model-based + offline
  • OpenReview: 7, 6, 6, 5
  • ExpEnv: d4rl dataset

• Offline Model-based Adaptable Policy Learning

  • Xiong-Hui Chen, Yang Yu, Qingyang Li, Fan-Ming Luo, Zhiwei Tony Qin, Shang Wenjie, Jieping Ye
  • Key: model-based + offline
  • OpenReview: 6, 6, 6, 4
  • ExpEnv: d4rl dataset

• Weighted model estimation for offline model-based reinforcement learning

  • Toru Hishinuma, Kei Senda
  • Key: model-based + offline
  • OpenReview: 7, 6, 6, 6
  • ExpEnv: pendulum, d4rl dataset

• Reward-Free Model-Based Reinforcement Learning with Linear Function Approximation

  • Weitong Zhang, Dongruo Zhou, Quanquan Gu
  • Key: learning theory + model-based reward-free RL + linear function approximation
  • OpenReview: 6, 6, 5, 5
  • ExpEnv: None

• Provable Model-based Nonlinear Bandit and Reinforcement Learning: Shelve Optimism, Embrace Virtual Curvature

  • Kefan Dong, Jiaqi Yang, Tengyu Ma
  • Key: learning theory + model-based bandit RL + nonlinear function approximation
  • OpenReview: 7, 7, 7, 6
  • ExpEnv: None

ICLR 2021

• Deployment-Efficient Reinforcement Learning via Model-Based Offline Optimization

  • Tatsuya Matsushima, Hiroki Furuta, Yutaka Matsuo, Ofir Nachum, Shixiang Gu
  • Key: model-based + behavior cloning (warmup) + trpo
  • OpenReview: 8, 7, 7, 5
  • ExpEnv: d4rl dataset

• Control-Aware Representations for Model-based Reinforcement Learning

  • Brandon Cui, Yinlam Chow, Mohammad Ghavamzadeh
  • Key: representation learning + model-based soft actor-critic
  • OpenReview: 6, 6, 6
  • ExpEnv: planar system, inverted pendulum – swingup, cartpole, 3-link manipulator — swingUp & balance

• Mastering Atari with Discrete World Models

  • Danijar Hafner, Timothy Lillicrap, Mohammad Norouzi, Jimmy Ba
  • Key: Dreamer V1 + many tricks(multiple categorical variables, KL balancing, etc)
  • OpenReview: 9, 8, 5, 4
  • ExpEnv: atari

• Model-Based Visual Planning with Self-Supervised Functional Distances

  • Stephen Tian, Suraj Nair, Frederik Ebert, Sudeep Dasari, Benjamin Eysenbach, Chelsea Finn, Sergey Levine
  • Key: goal-reaching task + dynamics learning + distance learning (goal-conditioned Q-function)
  • OpenReview: 7, 7, 7, 7
  • ExpEnv: sawyer, door sliding

• Model-Based Offline Planning

  • Arthur Argenson, Gabriel Dulac-Arnold
  • Key: model-based + offline
  • OpenReview: 8, 7, 5, 5
  • ExpEnv: RL Unplugged(RLU), d4rl dataset

• Offline Model-Based Optimization via Normalized Maximum Likelihood Estimation

  • Justin Fu, Sergey Levine
  • Key: model-based + offline
  • OpenReview: 8, 6, 6
  • ExpEnv: design-bench

• On the role of planning in model-based deep reinforcement learning

  • Jessica B. Hamrick, Abram L. Friesen, Feryal Behbahani, Arthur Guez, Fabio Viola, Sims Witherspoon, Thomas Anthony, Lars Buesing, Petar Veličković, Théophane Weber
  • Key: discussion about planning in MuZero
  • OpenReview: 7, 7, 6, 5
  • ExpEnv: atari, go, deepmind control suite

• Representation Balancing Offline Model-based Reinforcement Learning

  • Byung-Jun Lee, Jongmin Lee, Kee-Eung Kim
  • Key: Representation Balancing MDP + model-based + offline
  • OpenReview: 7, 7, 7, 6
  • ExpEnv: d4rl dataset

• Model-based micro-data reinforcement learning: what are the crucial model properties and which model to choose?

  • Balázs Kégl, Gabriel Hurtado, Albert Thomas
  • Key: mixture density nets + heteroscedasticity
  • OpenReview: 7, 7, 7, 6, 5
  • ExpEnv: acrobot system

ICML 2021

• Conservative Objective Models for Effective Offline Model-Based Optimization

  • Brandon Trabucco, Aviral Kumar, Xinyang Geng, Sergey Levine
  • Key: conservative objective model + offline mbrl
  • ExpEnv: design-bench

• Continuous-Time Model-Based Reinforcement Learning

  • Çağatay Yıldız, Markus Heinonen, Harri Lähdesmäki
  • Key: continuous-time
  • ExpEnv: pendulum, cartPole and acrobot

• Model-Based Reinforcement Learning via Latent-Space Collocation

  • Oleh Rybkin, Chuning Zhu, Anusha Nagabandi, Kostas Daniilidis, Igor Mordatch, Sergey Levine
  • Key: latent space collocation
  • ExpEnv: sparse metaworld tasks

• Model-Free and Model-Based Policy Evaluation when Causality is Uncertain

  • David A Bruns-Smith
  • Key: worst-case bounds
  • ExpEnv: ope-tools

• Muesli: Combining Improvements in Policy Optimization

  • Matteo Hessel, Ivo Danihelka, Fabio Viola, Arthur Guez, Simon Schmitt, Laurent Sifre, Theophane Weber, David Silver, Hado van Hasselt
  • Key: value equivalence
  • ExpEnv: atari

• PC-MLP: Model-based Reinforcement Learning with Policy Cover Guided Exploration

  • Yuda Song, Wen Sun
  • Key: sample complexity + kernelized nonlinear regulators + linear MDPs
  • ExpEnv: mountain car, antmaze, mujoco

• Temporal Predictive Coding For Model-Based Planning In Latent Space

  • Tung Nguyen, Rui Shu, Tuan Pham, Hung Bui, Stefano Ermon
  • Key: temporal predictive coding with a RSSM + latent space
  • ExpEnv: deepmind control suite

• Model-based Reinforcement Learning for Continuous Control with Posterior Sampling

  • Ying Fan, Yifei Ming
  • Key: regret bound of psrl + mpc
  • ExpEnv: continuous cartpole, pendulum swingup,, mujoco

• A Sharp Analysis of Model-based Reinforcement Learning with Self-Play

  • Qinghua Liu, Tiancheng Yu, Yu Bai, Chi Jin
  • Key: learning theory + multi-agent + model-based self play + two-player zero-sum Markov games
  • ExpEnv: None

License

Awesome Model-Based RL is released under the Apache 2.0 license.