# Robust Imitation of a Few Demonstrations with a Backwards Model

arXiv: https://arxiv.org/abs/2210.09337 · HTML: https://mlsanity.com/paper/2210.09337/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 61 references that are in the map.
- First posted on arXiv 17 Oct 2022.

## Builds on (most cited first, 30 of 31)

- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Conservative Q-Learning for Offline Reinforcement Learning](https://mlsanity.com/paper/2006.04779/) (2020)
- [Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1606.03476/) (2016)
- [D4RL: Datasets for Deep Data-Driven Reinforcement Learning](https://mlsanity.com/paper/2004.07219/) (2020)
- [Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model](https://mlsanity.com/paper/1911.08265/) (2019)
- [Off-Policy Deep Reinforcement Learning without Exploration](https://mlsanity.com/paper/1812.02900/) (2018)
- [When to Trust Your Model: Model-Based Policy Optimization](https://mlsanity.com/paper/1906.08253/) (2019)
- [Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations](https://mlsanity.com/paper/1709.10087/) (2017)
- [An Algorithmic Perspective on Imitation Learning](https://mlsanity.com/paper/1811.06711/) (2018)
- [Overcoming Exploration in Reinforcement Learning with Demonstrations](https://mlsanity.com/paper/1709.10089/) (2017)
- [Leveraging Demonstrations for Deep Reinforcement Learning on Robotics Problems with Sparse Rewards](https://mlsanity.com/paper/1707.08817/) (2017)
- [Multi-Goal Reinforcement Learning: Challenging Robotics Environments and Request for Research](https://mlsanity.com/paper/1802.09464/) (2018)
- [How Neural Networks Extrapolate: From Feedforward to Graph Neural Networks](https://mlsanity.com/paper/2009.11848/) (2020)
- [DART: Noise Injection for Robust Imitation Learning](https://mlsanity.com/paper/1703.09327/) (2017)
- [Planning with Goal-Conditioned Policies](https://mlsanity.com/paper/1911.08453/) (2019)
- [Reinforcement and Imitation Learning for Diverse Visuomotor Skills](https://mlsanity.com/paper/1802.09564/) (2018)
- [Context-aware Dynamics Model for Generalization in Model-Based Reinforcement Learning](https://mlsanity.com/paper/2005.06800/) (2020)
- [Deeply AggreVaTeD: Differentiable Imitation Learning for Sequential Prediction](https://mlsanity.com/paper/1703.01030/) (2017)
- [Robust Imitation of Diverse Behaviors](https://mlsanity.com/paper/1707.02747/) (2017)
- [Offline Reinforcement Learning with Reverse Model-based Imagination](https://mlsanity.com/paper/2110.00188/) (2021)
- [PlayVirtual: Augmenting Cycle-Consistent Virtual Trajectories for Reinforcement Learning](https://mlsanity.com/paper/2106.04152/) (2021)
- [Bidirectional Model-based Policy Optimization](https://mlsanity.com/paper/2007.01995/) (2020)
- [Constrained Markov Decision Processes via Backward Value Functions](https://mlsanity.com/paper/2008.11811/) (2020)
- [Funnel Libraries for Real-Time Robust Feedback Motion Planning](https://mlsanity.com/paper/1601.04037/) (2016)
- [Recall Traces: Backtracking Models for Efficient Reinforcement Learning](https://mlsanity.com/paper/1804.00379/) (2018)
- [Forward-Backward Reinforcement Learning](https://mlsanity.com/paper/1803.10227/) (2018)
- [There Is No Turning Back: A Self-Supervised Approach for Reversibility-Aware Reinforcement Learning](https://mlsanity.com/paper/2106.04480/) (2021)
- [Generative predecessor models for sample-efficient imitation learning](https://mlsanity.com/paper/1904.01139/) (2019)
- [Learning Self-Correctable Policies and Value Functions from Demonstrations with Negative Sampling](https://mlsanity.com/paper/1907.05634/) (2019)
- [Learning Retrospective Knowledge with Reverse Reinforcement Learning](https://mlsanity.com/paper/2007.06703/) (2020)

## Built on by (most cited first, 7 of 7)

- [Diffusion Meets DAgger: Supercharging Eye-in-hand Imitation Learning](https://mlsanity.com/paper/2402.17768/) (2024)
- [CCIL: Continuity-based Data Augmentation for Corrective Imitation Learning](https://mlsanity.com/paper/2310.12972/) (2023)
- [From Past to Future: Rethinking Eligibility Traces](https://mlsanity.com/paper/2312.12972/) (2023)
- [Sample-Efficient Behavior Cloning Using General Domain Knowledge](https://mlsanity.com/paper/2501.16546/) (2025)
- [GuidedAttention: Interpretable and Correctable Visual Attention for OOD-Robust Robot Manipulation via Imitation Learning](https://mlsanity.com/paper/2607.21049/) (2026)
- [Mitigating State Aliasing in Vision-Language-Action Models via Inverse Dynamics Learning](https://mlsanity.com/paper/2605.29577/) (2026)
- [Beyond Conservatism: Recoverability-Conditioned Exploration for Model-Based Imitation Learning](https://mlsanity.com/paper/2609.33336/) (2026)

## Abstract (arXiv)

Behavior cloning of expert demonstrations can speed up learning optimal policies in a more sample-efficient way over reinforcement learning. However, the policy cannot extrapolate well to unseen states outside of the demonstration data, creating covariate shift (agent drifting away from demonstrations) and compounding errors. In this work, we tackle this issue by extending the region of attraction around the demonstrations so that the agent can learn how to get back onto the demonstrated trajectories if it veers off-course. We train a generative backwards dynamics model and generate short imagined trajectories from states in the demonstrations. By imitating both demonstrations and these model rollouts, the agent learns the demonstrated paths and how to get back onto these paths. With optimal or near-optimal demonstrations, the learned policy will be both optimal and robust to deviations, with a wider region of attraction. On continuous control domains, we evaluate the robustness when starting from different initial states unseen in the demonstration data. While both our method and other imitation learning baselines can successfully solve the tasks for initial states in the training distribution, our method exhibits considerably more robustness to different initial states.