mlsanityData to 14 Oct 2026

UniIntervene++: An Adaptive Intervention Agent for Efficient Real-World Reinforcement Learning

arXiv 2610.03620 ↗First posted 2 Oct 2026cs.LG cs.ROcited by 0

On the map: Deep reinforcement learning methods → Deep RL algorithms and representations also fits: RL for robot locomotion and manipulation

Abstract (arXiv)

Online reinforcement learning (RL) enables robot policies to improve through physical interaction, but the assistance they require changes as their competence evolves. Existing intervention strategies based on offline estimates or fixed decision rules can therefore become mismatched to the current policy. To address this, we propose UniIntervene++, an adaptive intervention agent that learns to allocate control between autonomous execution and heterogeneous assisted behaviors during online RL. Specifically, UniIntervene++ first formulates the evolving RL policy, trajectory correction, and a task-structured CodePolicy as Options in a unified semi-Markov decision process and learns their relative values online. Building on this, competence-adaptive intervention periodically probes the RL policy through unassisted execution, keeping control allocation responsive to its evolving capability. Finally, coupled experience learning allows assisted behaviors to improve the RL policy, whose evolving outcomes in turn reshape future intervention decisions. In this way, UniIntervene++ jointly determines when to intervene, how to intervene, and when to return control as the RL policy improves. Across five real-world manipulation tasks, UniIntervene++ achieves an average success rate of 89.67%, outperforming all baselines by at least 6 percentage points, while reducing human intervention to 0.77%, a relative reduction of at least 94.6% from the best baseline. Code is available in our hrefhttps://github.com/dannyyudong/An-Adaptive-Intervention-Agent-for-Efficient-Real-World-Reinforcement-LearningGitHub repository.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Deep RL algorithms and representations, one of 15 directions in Deep reinforcement learning methods. Builds on 25 of its 33 references that are in the map. First posted on arXiv 2 Oct 2026.

Builds on

25 of 33 references.

References · 25

  1. 2015Deep Reinforcement Learning with Double Q-learningcited by 1,745
  2. 2023Efficient Online Reinforcement Learning with Offline Datacited by 363
  3. 2018HG-DAgger: Interactive Imitation Learning with Human Expertscited by 266
  4. 2024Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learningcited by 189
  5. 2020Recovery RL: Safe Reinforcement Learning with Learned Recovery Zonescited by 175
  6. 2024SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learningcited by 136
  7. 2021ThriftyDAgger: Budget-Aware Novelty and Risk Gating for Interactive Imitation Learningcited by 108
  8. 2022Robot Learning on the Job: Human-in-the-Loop Autonomy and Learning During Deploymentcited by 108
  9. 2025RL-100: Performant Robotic Manipulation with Real-World Reinforcement Learningcited by 59
  10. 2023Self-Improving Robots: End-to-End Autonomous Visuomotor Reinforcement Learningcited by 34
  11. 2023RLIF: Interactive Imitation Learning as Reinforcement Learningcited by 30
  12. 2022Back to the Manifold: Recovering from Out-of-Distribution Statescited by 20
13 more
  1. 2026Failure-Aware RL: Reliable Offline-to-Online Reinforcement Learning with Self-Recovery for Real-World Manipulationcited by 11
  2. 2021LS3: Latent Space Safe Sets for Long-Horizon Visuomotor Control of Sparse Reward Iterative Taskscited by 10
  3. 2024RecoveryChaining: Learning Local Recovery Policies for Robust Manipulationcited by 9
  4. 2022Efficient Recovery Learning using Model Predictive Meta-Reasoningcited by 8
  5. 2025Uncertainty Comes for Free: Human-in-the-Loop Policies with Diffusion Modelscited by 8
  6. 2025Real-world Reinforcement Learning from Suboptimal Interventionscited by 8
  7. 2026UniIntervene: Agentic Intervention for Efficient Real-World Reinforcement Learningcited by 6
  8. 2026E2HiL: Entropy-Guided Sample Selection for Efficient Real-World Human-in-the-Loop Reinforcement Learningcited by 5
  9. 2020Learning Skills to Patch Plans Based on Inaccurate Modelscited by 3
  10. 2026OHP-RL: Online Human Preference as Guidance in Reinforcement Learning for Robot Manipulationcited by 3
  11. 2026WorldSample: Closed-loop Real-robot RL with World Modellingcited by 3
  12. 2026Facet-0: A Robotic Foundation Model for Contact-Rich Precise Manipulationcited by 2
  13. 2026One Demonstration Is Enough for Real-World Robotic Reinforcement Learningcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2023H2O+: An Improved Framework for Hybrid Offline-and-Online RL with Dynamics Gaps
  2. 2026BEE: Intervention-Adaptive Real-World Reinforcement Learning with Vision-Language-Action Models
  3. 2021Efficient Robotic Manipulation Through Offline-to-Online Reinforcement Learning and Goal-Aware State Information
  4. 2026From Pretraining to Proficiency: Real-World Subtask RL for Long-Horizon Manipulation with Minimal Human Intervention
  5. 2026Self-adapting Robotic Agents through Online Continual Reinforcement Learning with World Model Feedback
  6. 2026SmoothRL: Online Reinforcement Learning During Asynchronous Execution
  7. 2026Support-Constrained RL Enables Real-World Policy Improvement without Real-World Experience
  8. 2026Evo-RL: Iterative Robot Reinforcement Learning from Selective Experience