The MAGICAL Benchmark for Robust Imitation

Abstract

Imitation Learning (IL) algorithms are typically evaluated in the same environment that was used to create demonstrations. This rewards precise reproduction of demonstrations in one particular environment, but provides little information about how robustly an algorithm can generalise the demonstrator's intent to substantially different deployment settings. This paper presents the MAGICAL benchmark suite, which permits systematic evaluation of generalisation by quantifying robustness to different kinds of distribution shift that an IL algorithm is likely to encounter in practice. Using the MAGICAL suite, we confirm that existing IL algorithms overfit significantly to the context in which demonstrations are provided. We also show that standard methods for reducing overfitting are effective at creating narrow perceptual invariances, but are not sufficient to enable transfer to contexts that require substantially different behaviour, which suggests that new approaches will be needed in order to robustly generalise demonstrator intent. Code and data for the MAGICAL suite is available at https://github.com/qxcv/magical/.

NeurIPS 2020 conference paper (poster)

References in corpus (14)

• Solving Rubik's Cube with a Robot Hand
• Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning
• Quantifying Generalization in Reinforcement Learning
• Leveraging Procedural Generation to Benchmark Reinforcement Learning
• Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow
• Gotta Learn Fast: A New Benchmark for Generalization in RL
• Bayesian multitask inverse reinforcement learning
• Discriminator-Actor-Critic: Addressing Sample Inefficiency and Reward Bias in Adversarial Imitation Learning
• rlpyt: A Research Code Base for Deep Reinforcement Learning in PyTorch
• Safe Imitation Learning via Fast Bayesian Reward Inference from Preferences
• Meta-Inverse Reinforcement Learning with Probabilistic Context Variables
• Task-Embedded Control Networks for Few-Shot Imitation Learning
• Multi-task Maximum Entropy Inverse Reinforcement Learning
• On the Measure of Intelligence

Cited by in corpus (3)

• What Matters in Learning from Offline Human Demonstrations for Robot Manipulation
• XIRL: Cross-embodiment Inverse Reinforcement Learning
• Learning dissection trajectories from expert surgical videos via imitation learning with equivariant diffusion