[2507.09061] Action Chunking and Exploratory Data Collection Yield Exponential Improvements in Behavior Cloning for Continuous Control

Abstract:This paper presents a theoretical analysis of two of the most impactful interventions in modern learning from demonstration in robotics and continuous control: the practice of action-chunking (predicting sequences of actions in open-loop) and exploratory augmentation of expert demonstrations. Though recent results show that learning from demonstration, also known as imitation learning (IL), can suffer errors that compound exponentially with task horizon in continuous settings, we demonstrate that action chunking and exploratory data collection circumvent exponential compounding errors in different regimes. Our results identify control-theoretic stability as the key mechanism underlying the benefits of these interventions. On the empirical side, we validate our predictions and the role of control-theoretic stability through experimentation on popular robot learning benchmarks. On the theoretical side, we demonstrate that the control-theoretic lens provides fine-grained insights into how compounding error arises, leading to tighter statistical guarantees on imitation learning error when these interventions are applied than previous techniques based on information-theoretic considerations alone.

Action Chunking and Exploratory Data Collection Yield Exponential Improvements in Behavior Cloning for Continuous Control This paper presents a theoretical analysis of two of the most impactful interventions in modern learning from demonstration in robotics and continuous control: the practice of action-chunking (predicting sequences of actions in open-loop) and exploratory augmentation of expert demonstrations. Though recent results show that learning from demonstration, also known as imitation learning (IL), can suffer errors that compound exponentially with task horizon in continuous setti

saved by

• Pranav Jadhav

related reading

• State of Robot Learning, December 2025vedder.io
• Ch. 21 - Imitation Learningunderactuated.mit.edu
• The Pitfalls of Imitation Learning when Actions are Continuousarxiv.org
• Reinforcement Learning via Implicit Imitation Guidancearxiv.org
• Learning Beyond Gradientstrinkle23897.github.io
• Causal Video Models Are Data-Efficient Robot Policy Learners | Rhoda AIrhoda.ai
• [1811.01848] Plan Online, Learn Offline: Efficient Learning and Exploration via Model-Based Controlarxiv.org
• Should I Use Offline RL or Imitation Learning? – The Berkeley Artificial Intelligence Research Blogbair.berkeley.edu
• Reinforcement Learning with Action Chunkingarxiv.org
• Learning to Imitate | SAIL Blogai.stanford.edu
• [2507.09041] Behavioral Exploration: Learning to Explore via In-Context Adaptationar5iv.labs.arxiv.org
• Learning Latent Plans from Playlearning-from-play.github.io