Reinforcement Learning in Robotics

Últimos artículos

• A Distributional Treatment of Real2Sim2Real for Object-Centric Agent Adaptation in Vision-Driven Deformable Linear Object Manipulation

  Georgios Kamaras, Subramanian Ramamoorthy · 26 de febrero de 2026

  We present an integrated (or end-to-end) framework for the Real2Sim2Real problem of manipulating deformable linear objects (DLOs) based on visual perception. Working with a parameterised set of DLOs, we use likelihood-free inference (LFI) to compute the posterior distributions for the physical param…

• Group Orthogonalized Policy Optimization:Group Policy Optimization as Orthogonal Projection in Hilbert Space

  Wang Zixian · 26 de febrero de 2026

  We present Group Orthogonalized Policy Optimization (GOPO), a new alignment algorithm for large language models derived from the geometry of Hilbert function spaces. Instead of optimizing on the probability simplex and inheriting the exponential curvature of Kullback-Leibler divergence, GOPO lifts a…

• Object-Centric World Models from Few-Shot Annotations for Sample-Efficient Reinforcement Learning

  Weipu Zhang, Adam Jelley, Trevor McInroe, Amos Storkey, Gang Wang · 26 de febrero de 2026

  While deep reinforcement learning (RL) from pixels has achieved remarkable success, its sample inefficiency remains a critical limitation for real-world applications. Model-based RL (MBRL) addresses this by learning a world model to generate simulated experience, but standard approaches that rely on…

• ARLArena: A Unified Framework for Stable Agentic Reinforcement Learning

  Xiaoxuan Wang, Han Zhang, Haixin Wang, Yidan Shi, Ruoyan Li, Kaiqiao Han, Chenyi Tong, Haoran Deng, Renliang Sun, Alexander Taylor, Yanqiao Zhu, Jason Cong, Yizhou Sun, Wei Wang · 26 de febrero de 2026

  Agentic reinforcement learning (ARL) has rapidly gained attention as a promising paradigm for training agents to solve complex, multi-step interactive tasks. Despite encouraging early results, ARL remains highly unstable, often leading to training collapse. This instability limits scalability to lar…

• Hierarchical Lead Critic based Multi-Agent Reinforcement Learning

  David Eckel, Henri Mee{\ss} · 26 de febrero de 2026

  Cooperative Multi-Agent Reinforcement Learning (MARL) solves complex tasks that require coordination from multiple agents, but is often limited to either local (independent learning) or global (centralized learning) perspectives. In this paper, we introduce a novel sequential training scheme and MAR…

• Training Generalizable Collaborative Agents via Strategic Risk Aversion

  Chengrui Qu, Yizhou Zhang, Nicholas Lanzetti, Eric Mazumdar · 26 de febrero de 2026

  Many emerging agentic paradigms require agents to collaborate with one another (or people) to achieve shared goals. Unfortunately, existing approaches to learning policies for such collaborative problems produce brittle solutions that fail when paired with new partners. We attribute these failures t…

• Generalisation of RLHF under Reward Shift and Clipped KL Regularisation

  Kenton Tang, Yuzhu Chen, Fengxiang He · 26 de febrero de 2026

  Alignment and adaptation in large language models heavily rely on reinforcement learning from human feedback (RLHF); yet, theoretical understanding of its generalisability remains premature, especially when the learned reward could shift, and the KL control is estimated and clipped. To address this …

• Tool-R0: Self-Evolving LLM Agents for Tool-Learning from Zero Data

  Emre Can Acikgoz, Cheng Qian, Jonas H\"ubotter, Heng Ji, Dilek Hakkani-T\"ur, Gokhan Tur · 26 de febrero de 2026

  Large language models (LLMs) are becoming the foundation for autonomous agents that can use tools to solve complex tasks. Reinforcement learning (RL) has emerged as a common approach for injecting such agentic capabilities, but typically under tightly controlled training setups. It often depends on …

• On the Structural Non-Preservation of Epistemic Behaviour under Policy Transformation

  Alexander Galozy · 26 de febrero de 2026

  Reinforcement learning (RL) agents under partial observability often condition actions on internally accumulated information such as memory or inferred latent context. We formalise such information-conditioned interaction patterns as behavioural dependency: variation in action selection with respect…

• Self-Correcting VLA: Online Action Refinement via Sparse World Imagination

  Chenyv Liu, Wentao Tan, Lei Zhu, Fengling Li, Jingjing Li, Guoli Yang, Heng Tao Shen · 26 de febrero de 2026

  Standard vision-language-action (VLA) models rely on fitting statistical data priors, limiting their robust understanding of underlying physical dynamics. Reinforcement learning enhances physical grounding through exploration yet typically relies on external reward signals that remain isolated from …

• Evolutionary System Prompt Learning for Reinforcement Learning in LLMs

  Lunjun Zhang, Ryan Chen, Bradly C. Stadie · 26 de febrero de 2026

  Building agentic systems that can autonomously self-improve from experience is a longstanding goal of AI. Large language models (LLMs) today primarily self-improve via two mechanisms: self-reflection for context updates, and reinforcement learning (RL) for weight updates. In this work, we propose Ev…

• Cooperative-Competitive Team Play of Real-World Craft Robots

  Rui Zhao, Xihui Li, Yizheng Zhang, Yuzhen Liu, Zhong Zhang, Yufeng Zhang, Cheng Zhou, Zhengyou Zhang, Lei Han · 25 de febrero de 2026

  Multi-agent deep Reinforcement Learning (RL) has made significant progress in developing intelligent game-playing agents in recent years. However, the efficient training of collective robots using multi-agent RL and the transfer of learned policies to real-world applications remain open research que…

• From Parameters to Behaviors: Unsupervised Compression of the Policy Space

  Davide Tenedini, Riccardo Zamboni, Mirco Mutti, Marcello Restelli · 25 de febrero de 2026

  Despite its recent successes, Deep Reinforcement Learning (DRL) is notoriously sample-inefficient. We argue that this inefficiency stems from the standard practice of optimizing policies directly in the high-dimensional and highly redundant parameter space $\Theta$. This challenge is greatly compoun…

• Balancing Multiple Objectives in Urban Traffic Control with Reinforcement Learning from AI Feedback

  Chenyang Zhao, Vinny Cahill, Ivana Dusparic · 25 de febrero de 2026

  Reward design has been one of the central challenges for real world reinforcement learning (RL) deployment, especially in settings with multiple objectives. Preference-based RL offers an appealing alternative by learning from human preferences over pairs of behavioural outcomes. More recently, RL fr…

• What Matters for Simulation to Online Reinforcement Learning on Real Robots

  Yarden As, Dhruva Tirumala, Ren\'e Zurbr\"ugg, Chenhao Li, Stelian Coros, Andreas Krause, Markus Wulfmeier · 25 de febrero de 2026

  We investigate what specific design choices enable successful online reinforcement learning (RL) on physical robots. Across 100 real-world training runs on three distinct robotic platforms, we systematically ablate algorithmic, systems, and experimental decisions that are typically left implicit in …

• Repurposing Synthetic Data for Fine-grained Search Agent Supervision

  Yida Zhao, Kuan Li, Xixi Wu, Liwen Zhang, Dingchu Zhang, Baixuan Li, Maojia Song, Zhuo Chen, Chenxi Wang, Xinyu Wang, Kewei Tu, Pengjun Xie, Jingren Zhou, Yong Jiang · 25 de febrero de 2026

  LLM-based search agents are increasingly trained on entity-centric synthetic data to solve complex, knowledge-intensive tasks. However, prevailing training methods like Group Relative Policy Optimization (GRPO) discard this rich entity information, relying instead on sparse, outcome-based rewards. T…

• $\kappa$-Explorer: A Unified Framework for Active Model Estimation in MDPs

  Xihe Gu, Urbashi Mitra, Tara Javidi · 25 de febrero de 2026

  In tabular Markov decision processes (MDPs) with perfect state observability, each trajectory provides active samples from the transition distributions conditioned on state-action pairs. Consequently, accurate model estimation depends on how the exploration policy allocates visitation frequencies in…

• Wasserstein Barycenter Soft Actor-Critic

  Zahra Shahrooei, Ali Baheri · 25 de febrero de 2026

  Deep off-policy actor-critic algorithms have emerged as the leading framework for reinforcement learning in continuous control domains. However, most of these algorithms suffer from poor sample efficiency, especially in environments with sparse rewards. In this paper, we take a step towards addressi…

• Performance Asymmetry in Model-Based Reinforcement Learning

  Jing Yu Lim, Rushi Shah, Zarif Ikram, Samson Yu, Haozhe Ma, Tze-Yun Leong, Dianbo Liu · 25 de febrero de 2026

  Recently, Model-Based Reinforcement Learning (MBRL) have achieved super-human level performance on the Atari100k benchmark on average. However, we discover that conventional aggregates mask a major problem, Performance Asymmetry: MBRL agents dramatically outperform humans in certain tasks (Agent-Opt…

• Gap-Dependent Bounds for Nearly Minimax Optimal Reinforcement Learning with Linear Function Approximation

  Haochen Zhang, Zhong Zheng, Lingzhou Xue · 25 de febrero de 2026

  We study gap-dependent performance guarantees for nearly minimax-optimal algorithms in reinforcement learning with linear function approximation. While prior works have established gap-dependent regret bounds in this setting, existing analyses do not apply to algorithms that achieve the nearly minim…

• Fuz-RL: A Fuzzy-Guided Robust Framework for Safe Reinforcement Learning under Uncertainty

  Xu Wan, Chao Yang, Cheng Yang, Jie Song, Mingyang Sun · 25 de febrero de 2026

  Safe Reinforcement Learning (RL) is crucial for achieving high performance while ensuring safety in real-world applications. However, the complex interplay of multiple uncertainty sources in real environments poses significant challenges for interpretable risk assessment and robust decision-making. …

• POMDPPlanners: Open-Source Package for POMDP Planning

  Yaacov Pariente, Vadim Indelman · 25 de febrero de 2026

  We present POMDPPlanners, an open-source Python package for empirical evaluation of Partially Observable Markov Decision Process (POMDP) planning algorithms. The package integrates state-of-the-art planning algorithms, a suite of benchmark environments with safety-critical variants, automated hyperp…

• Squint: Fast Visual Reinforcement Learning for Sim-to-Real Robotics

  Abdulaziz Almuzairee, Henrik I. Christensen · 25 de febrero de 2026

  Visual reinforcement learning is appealing for robotics but expensive -- off-policy methods are sample-efficient yet slow; on-policy methods parallelize well but waste samples. Recent work has shown that off-policy methods can train faster than on-policy methods in wall-clock time for state-based co…

• Matching Multiple Experts: On the Exploitability of Multi-Agent Imitation Learning

  Antoine Bergerault, Volkan Cevher, Negar Mehr · 25 de febrero de 2026

  Multi-agent imitation learning (MA-IL) aims to learn optimal policies from expert demonstrations of interactions in multi-agent interactive domains. Despite existing guarantees on the performance of the resulting learned policies, characterizations of how far the learned polices are from a Nash equi…

• Regret-Guided Search Control for Efficient Learning in AlphaZero

  Yun-Jui Tsai, Wei-Yu Chen, Yan-Ru Ju, Yu-Hung Chang, Ti-Rong Wu · 25 de febrero de 2026

  Reinforcement learning (RL) agents achieve remarkable performance but remain far less learning-efficient than humans. While RL agents require extensive self-play games to extract useful signals, humans often need only a few games, improving rapidly by repeatedly revisiting states where mistakes occu…