Learning from Suboptimal Demonstration via Self-Supervised Reward Regression

2020/10/17 by Letian Chen, Chen, Letian, Rohan Paleja +3 · 8 citations
Computer Science · #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning and Algorithms #Machine Learning and Data Classification #Reinforcement Learning in Robotics #Robotics (cs.RO)

paper · pdf · doi:10.48550/arxiv.2010.11723

openalex publication_date 2020/10/17 · openalex created_date 2020/10/29 · openalex updated_date 2026/07/28

Abstract

Learning from Demonstration (LfD) seeks to democratize robotics by enabling non-roboticist end-users to teach robots to perform a task by providing a human demonstration. However, modern LfD techniques, e.g. inverse reinforcement learning (IRL), assume users provide at least stochastically optimal demonstrations. This assumption fails to hold in most real-world scenarios. Recent attempts to learn from sub-optimal demonstration leverage pairwise rankings and following the Luce-Shepard rule. However, we show these approaches make incorrect assumptions and thus suffer from brittle, degraded performance. We overcome these limitations in developing a novel approach that bootstraps off suboptimal demonstrations to synthesize optimality-parameterized data to train an idealized reward function. We empirically validate we learn an idealized reward function with ~0.95 correlation with ground-truth reward versus ~0.75 for prior work. We can then train policies achieving ~200% improvement over the suboptimal demonstration and ~90% improvement over prior work. We present a physical demonstration of teaching a robot a topspin strike in table tennis that achieves 32% faster returns and 40% more topspin than user demonstration.

Citations

• Extrapolating Beyond Suboptimal Demonstrations via Inverse Reinforcement Learning from Observations
• A Reduction of Imitation Learning and Structured Prediction to No-Regret\n Online Learning

Cited by

• Dataset Poisoning Attacks on Behavioral Cloning Policies
• Robot Learning from a Physical World Model
• Reinforcement Fine-Tuning of Flow-Matching Policies for Vision-Language-Action Models
• When a Robot is More Capable than a Human: Learning from Constrained Demonstrators
• Learning from Failures: Understanding LLM Alignment through Failure-Aware Inverse RL
• Beyond Prediction: Reinforcement Learning as the Defining Leap in Healthcare AI
• Latent Policy Barrier: Learning Robust Visuomotor Policies by Staying In-Distribution
• A roadmap for AI in robotics

Related

• Towards Sample-efficient Apprenticeship Learning from Suboptimal Demonstration
• Reinforcement learning with Demonstrations from Mismatched Task under Sparse Reward
• Reward Machine Inference for Robotic Manipulation
• Extrapolating Beyond Suboptimal Demonstrations via Inverse Reinforcement Learning from Observations
• Fast Lifelong Adaptive Inverse Reinforcement Learning from Demonstrations