Reinforcement Learning in Robotics

Neueste Paper

• Learning with a Budget: Identifying the Best Arm with Resource Constraints

  Zitian Li, Wang Chi Cheung · 2. März 2026

  In many applications, evaluating the effectiveness of different alternatives comes with varying costs or resource usage. Motivated by such heterogeneity, we study the Best Arm Identification with Resource Constraints (BAIwRC) problem, where an agent seeks to identify the best alternative (aka arm) i…

• Foundation World Models for Agents that Learn, Verify, and Adapt Reliably Beyond Static Environments

  Florent Delgrange · 2. März 2026

  The next generation of autonomous agents must not only learn efficiently but also act reliably and adapt their behavior in open worlds. Standard approaches typically assume fixed tasks and environments with little or no novelty, which limits world models' ability to support agents that must evolve t…

• MAGE: Multi-scale Autoregressive Generation for Offline Reinforcement Learning

  Chenxing Lin, Xinhui Gao, Haipeng Zhang, Xinran Li, Haitao Wang, Songzhu Mei, Chenglu Wen, Weiquan Liu, Siqi Shen, Cheng Wang · 2. März 2026

  Generative models have gained significant traction in offline reinforcement learning (RL) due to their ability to model complex trajectory distributions. However, existing generation-based approaches still struggle with long-horizon tasks characterized by sparse rewards. Some hierarchical generation…

• Actor-Critic Pretraining for Proximal Policy Optimization

  Andreas Kernbach, Amr Elsheikh, Nicolas Grupp, Ren\'e Nagel, Marco F. Huber · 2. März 2026

  Reinforcement learning (RL) actor-critic algorithms enable autonomous learning but often require a large number of environment interactions, which limits their applicability in robotics. Leveraging expert data can reduce the number of required environment interactions. A common approach is actor pre…

• RUMAD: Reinforcement-Unifying Multi-Agent Debate

  Chao Wang, Han Lin, Huaze Tang, Huijing Lin, Wenbo Ding · 2. März 2026

  Multi-agent debate (MAD) systems leverage collective intelligence to enhance reasoning capabilities, yet existing approaches struggle to simultaneously optimize accuracy, consensus formation, and computational efficiency. Static topology methods lack adaptability to task complexity variations, while…

• On the Generalization of SFT: A Reinforcement Learning Perspective with Reward Rectification

  Yongliang Wu, Yizhou Zhou, Zhou Ziheng, Yingzhe Peng, Xinyu Ye, Xinting Hu, Wenbo Zhu, Lu Qi, Ming-Hsuan Yang, Xu Yang · 2. März 2026

  In this work, we present a simple yet theoretically motivated improvement to Supervised Fine-Tuning (SFT) for the Large Language Model (LLM), addressing its limited generalization compared to reinforcement learning (RL). Through mathematical analysis, we reveal that standard SFT gradients implicitly…

• Actor-Critic for Continuous Action Chunks: A Reinforcement Learning Framework for Long-Horizon Robotic Manipulation with Sparse Reward

  Jiarui Yang, Bin Zhu, Jingjing Chen, Yu-Gang Jiang · 2. März 2026

  Existing reinforcement learning (RL) methods struggle with long-horizon robotic manipulation tasks, particularly those involving sparse rewards. While action chunking is a promising paradigm for robotic manipulation, using RL to directly learn continuous action chunks in a stable and data-efficient …

• Beyond State-Wise Mirror Descent: Offline Policy Optimization with Parameteric Policies

  Xiang Li, Nan Jiang, Yuheng Zhang · 2. März 2026

  We investigate the theoretical aspects of offline reinforcement learning (RL) under general function approximation. While prior works (e.g., Xie et al., 2021) have established the theoretical foundations of learning a good policy from offline data via pessimism, existing algorithms that are computat…

• Offline-to-Online Multi-Agent Reinforcement Learning with Offline Value Function Memory and Sequential Exploration

  Hai Zhong, Xun Wang, Zhuoran Li, Longbo Huang · 2. März 2026

  Offline-to-Online Reinforcement Learning has emerged as a powerful paradigm, leveraging offline data for initialization and online fine-tuning to enhance both sample efficiency and performance. However, most existing research has focused on single-agent settings, with limited exploration of the mult…

• SafeGen-LLM: Enhancing Safety Generalization in Task Planning for Robotic Systems

  Jialiang Fan, Weizhe Xu, Mengyu Liu, Oleg Sokolsky, Insup Lee, Fangxin Kong · 2. März 2026

  Safety-critical task planning in robotic systems remains challenging: classical planners suffer from poor scalability, Reinforcement Learning (RL)-based methods generalize poorly, and base Large Language Models (LLMs) cannot guarantee safety. To address this gap, we propose safety-generalizable larg…

• Adversarial Fine-tuning in Offline-to-Online Reinforcement Learning for Robust Robot Control

  Shingo Ayabe, Hiroshi Kera, Kazuhiko Kawamoto · 2. März 2026

  Offline reinforcement learning enables sample-efficient policy acquisition without risky online interaction, yet policies trained on static datasets remain brittle under action-space perturbations such as actuator faults. This study introduces an offline-to-online framework that trains policies on c…

• DRL-ORA: Distributional Reinforcement Learning with Online Risk Adaption

  Yupeng Wu, Wenyun Li, Wenjie Huang, Chin Pang Ho · 2. März 2026

  One of the main challenges in reinforcement learning (RL) is that the agent has to make decisions that would influence the future performance without having complete knowledge of the environment. Dynamically adjusting the level of epistemic risk during the learning process can help to achieve reliab…

• Planning under Distribution Shifts with Causal POMDPs

  Matteo Ceriscioli, Karthika Mohan · 2. März 2026

  In the real world, planning is often challenged by distribution shifts. As such, a model of the environment obtained under one set of conditions may no longer remain valid as the distribution of states or the environment dynamics change, which in turn causes previously learned strategies to fail. In…

• Resilient Strategies for Stochastic Systems: How Much Does It Take to Break a Winning Strategy?

  Kush Grover, Markel Zubia, Debraj Chakraborty, Muqsit Azeem, Nils Jansen, Jan Kretinsky · 2. März 2026

  We study the problem of resilient strategies in the presence of uncertainty. Resilient strategies enable an agent to make decisions that are robust against disturbances. In particular, we are interested in those disturbances that are able to flip a decision made by the agent. Such a disturbance may,…

• Pessimistic Auxiliary Policy for Offline Reinforcement Learning

  Fan Zhang, Baoru Huang, Xin Zhang · 2. März 2026

  Offline reinforcement learning aims to learn an agent from pre-collected datasets, avoiding unsafe and inefficient real-time interaction. However, inevitable access to out-ofdistribution actions during the learning process introduces approximation errors, causing the error accumulation and considera…

• Learning to maintain safety through expert demonstrations in settings with unknown constraints: A Q-learning perspective

  George Papadopoulos, George A. Vouros · 2. März 2026

  Given a set of trajectories demonstrating the execution of a task safely in a constrained MDP with observable rewards but with unknown constraints and non-observable costs, we aim to find a policy that maximizes the likelihood of demonstrated trajectories trading the balance between being conservati…

• Bridging Dynamics Gaps via Diffusion Schr\"odinger Bridge for Cross-Domain Reinforcement Learning

  Hanping Zhang, Yuhong Guo · 2. März 2026

  Cross-domain reinforcement learning (RL) aims to learn transferable policies under dynamics shifts between source and target domains. A key challenge lies in the lack of target-domain environment interaction and reward supervision, which prevents direct policy learning. To address this challenge, we…

• Bridging the Performance Gap Between Target-Free and Target-Based Reinforcement Learning

  Th\'eo Vincent, Yogesh Tripathi, Tim Faust, Abdullah Akg\"ul, Yaniv Oren, Melih Kandemir, Jan Peters, Carlo D'Eramo · 2. März 2026

  The use of target networks in deep reinforcement learning is a widely popular solution to mitigate the brittleness of semi-gradient approaches and stabilize learning. However, target networks notoriously require additional memory and delay the propagation of Bellman updates compared to an ideal targ…

• Adaptive Correlation-Weighted Intrinsic Rewards for Reinforcement Learning

  Viet Bac Nguyen, Phuong Thai Nguyen · 2. März 2026

  We propose ACWI (Adaptive Correlation Weighted Intrinsic), an adaptive intrinsic reward scaling framework designed to dynamically balance intrinsic and extrinsic rewards for improved exploration in sparse reward reinforcement learning. Unlike conventional approaches that rely on manually tuned scala…

• Physics Informed Viscous Value Representations

  Hrishikesh Viswanath, Juanwu Lu, S. Talha Bukhari, Damon Conover, Ziran Wang, Aniket Bera · 27. Februar 2026

  Offline goal-conditioned reinforcement learning (GCRL) learns goal-conditioned policies from static pre-collected datasets. However, accurate value estimation remains a challenge due to the limited coverage of the state-action space. Recent physics-informed approaches have sought to address this by …

• Multi-agent imitation learning with function approximation: Linear Markov games and beyond

  Luca Viano, Till Freihaut, Emanuele Nevali, Volkan Cevher, Matthieu Geist, Giorgia Ramponi · 27. Februar 2026

  In this work, we present the first theoretical analysis of multi-agent imitation learning (MAIL) in linear Markov games where both the transition dynamics and each agent's reward function are linear in some given features. We demonstrate that by leveraging this structure, it is possible to replace t…

• When to Act, Ask, or Learn: Uncertainty-Aware Policy Steering

  Jessie Yuan, Yilin Wu, Andrea Bajcsy · 27. Februar 2026

  Policy steering is an emerging way to adapt robot behaviors at deployment-time: a learned verifier analyzes low-level action samples proposed by a pre-trained policy (e.g., diffusion policy) and selects only those aligned with the task. While Vision-Language Models (VLMs) are promising general-purpo…

• Accelerated Online Risk-Averse Policy Evaluation in POMDPs with Theoretical Guarantees and Novel CVaR Bounds

  Yaacov Pariente, Vadim Indelman · 27. Februar 2026

  Risk-averse decision-making under uncertainty in partially observable domains is a central challenge in artificial intelligence and is essential for developing reliable autonomous agents. The formal framework for such problems is the partially observable Markov decision process (POMDP), where risk s…

• Hierarchy-of-Groups Policy Optimization for Long-Horizon Agentic Tasks

  Shuo He, Lang Feng, Qi Wei, Xin Cheng, Lei Feng, Bo An · 27. Februar 2026

  Group-based reinforcement learning (RL), such as GRPO, has advanced the capabilities of large language models on long-horizon agentic tasks. To enable more fine-grained policy updates, recent research has increasingly shifted toward stepwise group-based policy optimization, which treats each step in…

• Mastering Multi-Drone Volleyball through Hierarchical Co-Self-Play Reinforcement Learning

  Ruize Zhang, Sirui Xiang, Zelai Xu, Feng Gao, Shilong Ji, Wenhao Tang, Wenbo Ding, Chao Yu, Yu Wang · 27. Februar 2026

  In this paper, we tackle the problem of learning to play 3v3 multi-drone volleyball, a new embodied competitive task that requires both high-level strategic coordination and low-level agile control. The task is turn-based, multi-agent, and physically grounded, posing significant challenges due to it…