Residual Off-Policy RL for Finetuning Behavior Cloning Policies

• URL: http://arxiv.org/abs/2509.19301v2
• Date: Thu, 25 Sep 2025 20:53:20 GMT
• Title: Residual Off-Policy RL for Finetuning Behavior Cloning Policies
• Authors: Lars Ankile, Zhenyu Jiang, Rocky Duan, Guanya Shi, Pieter Abbeel, Anusha Nagabandi,
• Abstract summary: We present a recipe that combines the benefits of behavior cloning (BC) and reinforcement learning (RL) through a residual learning framework.<n>Our method requires only sparse binary reward signals and can effectively improve manipulation policies on high-degree-of-freedom (DoF) systems.<n>In particular, we demonstrate, to the best of our knowledge, the first successful real-world RL training on a humanoid robot with dexterous hands.
• Score: 41.99435186991878
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advances in behavior cloning (BC) have enabled impressive visuomotor control policies. However, these approaches are limited by the quality of human demonstrations, the manual effort required for data collection, and the diminishing returns from offline data. In comparison, reinforcement learning (RL) trains an agent through autonomous interaction with the environment and has shown remarkable success in various domains. Still, training RL policies directly on real-world robots remains challenging due to sample inefficiency, safety concerns, and the difficulty of learning from sparse rewards for long-horizon tasks, especially for high-degree-of-freedom (DoF) systems. We present a recipe that combines the benefits of BC and RL through a residual learning framework. Our approach leverages BC policies as black-box bases and learns lightweight per-step residual corrections via sample-efficient off-policy RL. We demonstrate that our method requires only sparse binary reward signals and can effectively improve manipulation policies on high-degree-of-freedom (DoF) systems in both simulation and the real world. In particular, we demonstrate, to the best of our knowledge, the first successful real-world RL training on a humanoid robot with dexterous hands. Our results demonstrate state-of-the-art performance in various vision-based tasks, pointing towards a practical pathway for deploying RL in the real world.

Related papers

• Human-in-the-loop Online Rejection Sampling for Robotic Manipulation [55.99788088622936]
  Hi-ORS stabilizes value estimation by filtering out negatively rewarded samples during online fine-tuning.<n>Hi-ORS fine-tunes a pi-base policy to master contact-rich manipulation in just 1.5 hours of real-world training.
  arXiv  Detail & Related papers  (2025-10-30T11:53:08Z)
• Online Process Reward Leanring for Agentic Reinforcement Learning [92.26560379363492]
  Large language models (LLMs) are increasingly trained with reinforcement learning (RL) as autonomous agents.<n>Recent work attempts to integrate process supervision into agent learning but suffers from biased annotation.<n>We introduce Online Process Reward Learning (OPRL), a general credit-assignment strategy for agentic RL.
  arXiv  Detail & Related papers  (2025-09-23T16:15:42Z)
• Steering Your Diffusion Policy with Latent Space Reinforcement Learning [46.598122553180005]
  Behavioral cloning (BC)-learned policies typically require collecting additional human demonstrations to further improve their behavior.<n> reinforcement learning (RL) holds the promise of enabling autonomous online policy improvement, but often falls short of achieving this due to the large number of samples it typically requires.<n>We show that DSRL is highly sample efficient, requires only black-box access to the BC policy, and enables effective real-world autonomous policy improvement.
  arXiv  Detail & Related papers  (2025-06-18T18:35:57Z)
• Policy Agnostic RL: Offline RL and Online RL Fine-Tuning of Any Class and Backbone [72.17534881026995]
  We develop an offline and online fine-tuning approach called policy-agnostic RL (PA-RL)<n>We show the first result that successfully fine-tunes OpenVLA, a 7B generalist robot policy, autonomously with Cal-QL, an online RL fine-tuning algorithm.
  arXiv  Detail & Related papers  (2024-12-09T17:28:03Z)
• Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learning [47.785786984974855]
  We present a human-in-the-loop vision-based RL system that demonstrates impressive performance on a diverse set of dexterous manipulation tasks.<n>Our approach integrates demonstrations and human corrections, efficient RL algorithms, and other system-level design choices to learn policies.<n>We show that our method significantly outperforms imitation learning baselines and prior RL approaches, with an average 2x improvement in success rate and 1.8x faster execution.
  arXiv  Detail & Related papers  (2024-10-29T08:12:20Z)
• CIMRL: Combining IMitation and Reinforcement Learning for Safe Autonomous Driving [45.05135725542318]
  IMitation and Reinforcement Learning (CIMRL) approach enables training driving policies in simulation through leveraging imitative motion priors and safety constraints. By combining RL and imitation, we demonstrate our method achieves state-of-the-art results in closed loop simulation and real world driving benchmarks.
  arXiv  Detail & Related papers  (2024-06-13T07:31:29Z)
• Hybrid Inverse Reinforcement Learning [34.793570631021005]
  inverse reinforcement learning approach to imitation learning is a double-edged sword. We propose using hybrid RL -- training on a mixture of online and expert data -- to curtail unnecessary exploration. We derive both model-free and model-based hybrid inverse RL algorithms with strong policy performance guarantees.
  arXiv  Detail & Related papers  (2024-02-13T23:29:09Z)
• REBOOT: Reuse Data for Bootstrapping Efficient Real-World Dexterous Manipulation [61.7171775202833]
  We introduce an efficient system for learning dexterous manipulation skills withReinforcement learning. The main idea of our approach is the integration of recent advances in sample-efficient RL and replay buffer bootstrapping. Our system completes the real-world training cycle by incorporating learned resets via an imitation-based pickup policy.
  arXiv  Detail & Related papers  (2023-09-06T19:05:31Z)
• Jump-Start Reinforcement Learning [68.82380421479675]
  We present a meta algorithm that can use offline data, demonstrations, or a pre-existing policy to initialize an RL policy. In particular, we propose Jump-Start Reinforcement Learning (JSRL), an algorithm that employs two policies to solve tasks. We show via experiments that JSRL is able to significantly outperform existing imitation and reinforcement learning algorithms.
  arXiv  Detail & Related papers  (2022-04-05T17:25:22Z)
• AWAC: Accelerating Online Reinforcement Learning with Offline Datasets [84.94748183816547]
  We show that our method, advantage weighted actor critic (AWAC), enables rapid learning of skills with a combination of prior demonstration data and online experience. Our results show that incorporating prior data can reduce the time required to learn a range of robotic skills to practical time-scales.
  arXiv  Detail & Related papers  (2020-06-16T17:54:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.