Reinforcement Learning in Robotics

Neueste Paper

• Self-adapting Robotic Agents through Online Continual Reinforcement Learning with World Model Feedback

  Fabian Domberg, Georg Schildbach · 5. März 2026

  As learning-based robotic controllers are typically trained offline and deployed with fixed parameters, their ability to cope with unforeseen changes during operation is limited. Biologically inspired, this work presents a framework for online Continual Reinforcement Learning that enables automated …

• Learning Approximate Nash Equilibria in Cooperative Multi-Agent Reinforcement Learning via Mean-Field Subsampling

  Emile Anand, Ishani Karmarkar · 5. März 2026

  Many large-scale platforms and networked control systems have a centralized decision maker interacting with a massive population of agents under strict observability constraints. Motivated by such applications, we study a cooperative Markov game with a global agent and $n$ homogeneous local agents i…

• Unraveling the Complexity of Memory in RL Agents: an Approach for Classification and Evaluation

  Egor Cherepanov, Nikita Kachaev, Artem Zholus, Alexey K. Kovalev, Aleksandr I. Panov · 5. März 2026

  The incorporation of memory into agents is essential for numerous tasks within the domain of Reinforcement Learning (RL). In particular, memory is paramount for tasks that require the use of past information, adaptation to novel environments, and improved sample efficiency. However, the term "memory…

• Function Induction and Task Generalization: An Interpretability Study with Off-by-One Addition

  Qinyuan Ye, Robin Jia, Xiang Ren · 5. März 2026

  Large language models demonstrate the intriguing ability to perform unseen tasks via in-context learning. However, it remains unclear what mechanisms inside the model drive such task-level generalization. In this work, we approach this question through the lens of off-by-one addition (i.e., 1+1=3, 2…

• Minimax Optimal Strategy for Delayed Observations in Online Reinforcement Learning

  Harin Lee, Kevin Jamieson · 5. März 2026

  We study reinforcement learning with delayed state observation, where the agent observes the current state after some random number of time steps. We propose an algorithm that combines the augmentation method and the upper confidence bound approach. For tabular Markov decision processes (MDPs), we d…

• Q-Guided Stein Variational Model Predictive Control via RL-informed Policy Prior

  Shizhe Cai, Zeya Yin, Jayadeep Jacob, Fabio Ramos · 5. März 2026

  Model Predictive Control (MPC) enables reliable trajectory optimization under dynamics constraints, but often depends on accurate dynamics models and carefully hand-designed cost functions. Recent learning-based MPC methods aim to reduce these modeling and cost-design burdens by learning dynamics, p…

• What Does Flow Matching Bring To TD Learning?

  Bhavya Agrawalla, Michal Nauman, Aviral Kumar · 5. März 2026

  Recent work shows that flow matching can be effective for scalar Q-value function estimation in reinforcement learning (RL), but it remains unclear why or how this approach differs from standard critics. Contrary to conventional belief, we show that their success is not explained by distributional R…

• ToolRLA: Multiplicative Reward Decomposition for Tool-Integrated Agents

  Pengbo Liu · 5. März 2026

  Tool-integrated agents that interleave reasoning with API calls are promising for complex tasks, yet aligning them for high-stakes, domain-specific deployment remains challenging: existing reinforcement learning approaches rely on coarse binary rewards that cannot distinguish tool selection errors f…

• AutoQD: Automatic Discovery of Diverse Behaviors with Quality-Diversity Optimization

  Saeed Hedayatian, Stefanos Nikolaidis · 5. März 2026

  Quality-Diversity (QD) algorithms have shown remarkable success in discovering diverse, high-performing solutions, but rely heavily on hand-crafted behavioral descriptors that constrain exploration to predefined notions of diversity. Leveraging the equivalence between policies and occupancy measures…

• Memory, Benchmark & Robots: A Benchmark for Solving Complex Tasks with Reinforcement Learning

  Egor Cherepanov, Nikita Kachaev, Alexey K. Kovalev, Aleksandr I. Panov · 5. März 2026

  Memory is crucial for enabling agents to tackle complex tasks with temporal and spatial dependencies. While many reinforcement learning (RL) algorithms incorporate memory, the field lacks a universal benchmark to assess an agent's memory capabilities across diverse scenarios. This gap is particularl…

• Principled Learning-to-Communicate with Quasi-Classical Information Structures

  Xiangyu Liu, Haoyi You, Kaiqing Zhang · 5. März 2026

  Learning-to-communicate (LTC) in partially observable environments has received increasing attention in deep multi-agent reinforcement learning, where the control and communication strategies are jointly learned. Meanwhile, the impact of communication on decision-making has been extensively studied …

• Q-Measure-Learning for Continuous State RL: Efficient Implementation and Convergence

  Shengbo Wang · 5. März 2026

  We study reinforcement learning in infinite-horizon discounted Markov decision processes with continuous state spaces, where data are generated online from a single trajectory under a Markovian behavior policy. To avoid maintaining an infinite-dimensional, function-valued estimate, we propose the no…

• GIPO: Gaussian Importance Sampling Policy Optimization

  Chengxuan Lu, Zhenquan Zhang, Shukuan Wang, Qunzhi Lin, Baigui Sun, Yang Liu · 5. März 2026

  Post-training with reinforcement learning (RL) has recently shown strong promise for advancing multimodal agents beyond supervised imitation. However, RL remains limited by poor data efficiency, particularly in settings where interaction data are scarce and quickly become outdated. To address this c…

• HALyPO: Heterogeneous-Agent Lyapunov Policy Optimization for Human-Robot Collaboration

  Hao Zhang, Yaru Niu, Yikai Wang, Ding Zhao, H. Eric Tseng · 5. März 2026

  To improve generalization and resilience in human-robot collaboration (HRC), robots must handle the combinatorial diversity of human behaviors and contexts, motivating multi-agent reinforcement learning (MARL). However, inherent heterogeneity between robots and humans creates a rationality gap (RG) …

• Unbiased Dynamic Pruning for Efficient Group-Based Policy Optimization

  Haodong Zhu, Yangyang Ren, Yanjing Li, Mingbao Lin, Linlin Yang, Xuhui Liu, Xiantong Zhen, Haiguang Liu, Baochang Zhang · 5. März 2026

  Group Relative Policy Optimization (GRPO) effectively scales LLM reasoning but incurs prohibitive computational costs due to its extensive group-based sampling requirement. While recent selective data utilization methods can mitigate this overhead, they could induce estimation bias by altering the u…

• IPD: Boosting Sequential Policy with Imaginary Planning Distillation in Offline Reinforcement Learning

  Yihao Qin, Yuanfei Wang, Hang Zhou, Peiran Liu, Hao Dong, Yiding Ji · 5. März 2026

  Decision transformer based sequential policies have emerged as a powerful paradigm in offline reinforcement learning (RL), yet their efficacy remains constrained by the quality of static datasets and inherent architectural limitations. Specifically, these models often struggle to effectively integra…

• RoboCasa365: A Large-Scale Simulation Framework for Training and Benchmarking Generalist Robots

  Soroush Nasiriany, Sepehr Nasiriany, Abhiram Maddukuri, Yuke Zhu · 5. März 2026

  Recent advances in robot learning have accelerated progress toward generalist robots that can perform everyday tasks in human environments. Yet it remains difficult to gauge how close we are to this vision. The field lacks a reproducible, large-scale benchmark for systematic evaluation. To fill this…

• Next Embedding Prediction Makes World Models Stronger

  George Bredis, Nikita Balagansky, Daniil Gavrilov, Ruslan Rakhimov · 4. März 2026

  Capturing temporal dependencies is critical for model-based reinforcement learning (MBRL) in partially observable, high-dimensional domains. We introduce NE-Dreamer, a decoder-free MBRL agent that leverages a temporal transformer to predict next-step encoder embeddings from latent state sequences, d…

• A Covering Framework for Offline POMDPs Learning using Belief Space Metric

  Youheng Zhu, Yiping Lu · 4. März 2026

  In off policy evaluation (OPE) for partially observable Markov decision processes (POMDPs), an agent must infer hidden states from past observations, which exacerbates both the curse of horizon and the curse of memory in existing OPE methods. This paper introduces a novel covering analysis framework…

• LLMs for High-Frequency Decision-Making: Normalized Action Reward-Guided Consistency Policy Optimization

  Yang Zhao, Zihao Li, Zhiyu Jiang, Dandan Ma, Ganchao Liu, Wenzhe Zhao · 4. März 2026

  While Large Language Models (LLMs) form the cornerstone of sequential decision-making agent development, they have inherent limitations in high-frequency decision tasks. Existing research mainly focuses on discrete embodied decision scenarios with low-frequency and significant semantic differences i…

• On the Structural Limitations of Weight-Based Neural Adaptation and the Role of Reversible Behavioral Learning

  Pardhu Sri Rushi Varma Konduru · 4. März 2026

  Neural models are usually adapted through changes in parameters shared among model components via fine-tuning, alignment-based training, and reinforcement learning. These changes have been found effective in short-term optimization. However, they result in long-term alterations in the model's base b…

• Non-Rectangular Average-Reward Robust MDPs: Optimal Policies and Their Transient Values

  Shengbo Wang, Nian Si · 4. März 2026

  We study non-rectangular robust Markov decision processes under the average-reward criterion, where the ambiguity set couples transition probabilities across states and the adversary commits to a stationary kernel for the entire horizon. We show that any history-dependent policy achieving sublinear …

• Post Hoc Extraction of Pareto Fronts for Continuous Control

  Raghav Thakar, Gaurav Dixit, Kagan Tumer · 4. März 2026

  Agents in the real world must often balance multiple objectives, such as speed, stability, and energy efficiency in continuous control. To account for changing conditions and preferences, an agent must ideally learn a Pareto frontier of policies representing multiple optimal trade-offs. Recent advan…

• EvoSkill: Automated Skill Discovery for Multi-Agent Systems

  Salaheddin Alzubi, Noah Provenzano, Jaydon Bingham, Weiyuan Chen, Tu Vu · 4. März 2026

  Coding agents are increasingly used as general-purpose problem solvers, but their flexibility does not by itself confer the domain expertise needed for specialized tasks. Recent work addresses this through \textit{agent skills}: reusable workflows, and code, that augment agents with domain-specific …

• Real-Time Generative Policy via Langevin-Guided Flow Matching for Autonomous Driving

  Tianze Zhu, Yinuo Wang, Wenjun Zou, Tianyi Zhang, Likun Wang, Letian Tao, Feihong Zhang, Yao Lyu, Shengbo Eben Li · 4. März 2026

  Reinforcement learning (RL) is a fundamental methodology in autonomous driving systems, where generative policies exhibit considerable potential by leveraging their ability to model complex distributions to enhance exploration. However, their inherent high inference latency severely impedes their de…