Representation Learning for RL

RL: Reinforcement Learning

Representation learning for reinforcement learning studies how agents can turn observations and interaction histories into features that preserve what matters for predicting outcomes and choosing actions. Common approaches include predictive and contrastive latent models, successor and spectral representations, bisimulation metrics, and regularizers that encourage sparse, stable, or disentangled features. These methods support visual control, exploration, offline goal-conditioned navigation, world modeling, and transfer across tasks or environments. A central tension is that reusable features must discard irrelevant variation without losing details needed for control, and better-looking representations do not always improve downstream performance.

Latest papers

69 papers18 papers in the last four weeks, up 200% on the four weeks before. 0.1% of all new papers.

ProDVI: Programmatic Dynamics Priors for Value Network Initialization
Aug 6, 2026Xinwei Liu, Junyuan Liang, Jianting Zhang +1Reinforcement Learning LLM-Guided RL
1Sun Yat-sen University · 2Purdue University
Reinformed Dreamer: An Asymmetric World Model Efficiently Trained through Latent Guidance
Jul 28, 2026Gaspard Lambrechts, Adrien Bolland, Daniel Ebi +1Reinforcement Learning World Models
1McGill University, Canada · 3Université de Liège, Belgium · 4Karlsruhe Institute of Technology, Germany
Factorized Spectral Representations for Reinforcement Learning
Jul 15, 2026Junyi Wu, Dan LiTensor Decomposition Representation Learning for RL
Department of Industrial and Systems Engineering University of Washington Seattle, WA, USA
Vision-Based Dribbling for Humanoid Soccer via Privileged Representation Learning
Jul 14, 2026Flavio Maiorana, Valerio Spagnoli, Eugenio Bugli +5RL for Robotics Reinforcement Learning
Dept. of Computer, Control, and Management Engineering Sapienza University of Rome, Rome (Italy) · Institut de Robòtica i Informàtica Industrial (CSIC-UPC), C/ Llorens i Artigas 4-6, 08028, Barcelona, Spain · École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland
From Pixels to Temporal Correlations: Learning Informative Representations for Reinforcement Learning Pre-training
Jul 1, 2026Jinwen Wang, Youfang Lin, Xiaobo Hu +4Contrastive Learning Reinforcement Learning
School of Computer Science & Technology, Beijing Jiaotong University, Beijing, China · Beijing Key Laboratory of Traffic Data Mining and Embodied Intelligence, Beijing Jiaotong University, Beijing, China · Beijing Jiaotong University, Beijing, China
Task-Relevant Representation Decoupling for Visual Reinforcement Learning Generalization
Jul 1, 2026Jinwen Wang, Youfang Lin, Xiaobo Hu +4Representation Learning Representation Learning for RL
Beijing Key Laboratory of Traffic Data Mining and Embodied Intelligence, School of Computer Science & Technology, Beijing Jiaotong University, China · CSSC Intelligent Innovation Research Institute, China and Zhejiang University, China
Learning to Adapt: Representation-Based Reinforcement Learning for Multi-Task Skill Transfer
Jun 11, 2026Aryan Naveen, Haitong Ma, Haldun Balim +1RL for Robotics Reinforcement Learning
Massachusetts Institute of Technology · Harvard School of Engineering and Applied Sciences
Representation Learning Enables Scalable Multitask Deep Reinforcement Learning
Jun 4, 2026Johan Obando-Ceron, Lu Li, Scott Fujimoto +3Reinforcement Learning Representation Learning for RL
Mila – Québec AI Institute · Université de Montréal · McGill University +2
ConTraIRL: Factorized Contrastive Abstractions for Transferable IRL
Jun 2, 2026Yikang Gui, Bikramjit Banerjee, Prashant DoshiDisentangled Representation Learning Reinforcement Learning
School of Computing, University of Georgia · School of Computing Sciences & Computer Engineering, The University of Southern Mississippi
The Terminal Representation in Reinforcement Learning
May 29, 2026Amir Esterhuysen, Anders JonssonRepresentation Learning for RL Trajectory Representation Learning
Dept. Information and Communication Technologies Universitat Pompeu Fabra Barcelona, Spain
Learning to Perceive the World Through Control: Empowerment-Based Representation Learning
May 28, 2026Mahsa Bastankhah, Sophie Broderick, Benjamin EysenbachRepresentation Learning Reinforcement Learning
Princeton University, USA
Exploiting Local Dynamics Regularity for Reusable Skills in Offline Hierarchical RL
May 25, 2026Sarthak Dayal, Abhinav Peri, Carl Qi +4Hierarchical RL Offline RL
Department of Computer Science University of Texas at Austin · UT Austin · OpenAI
Learning in Low-Dimensional Subspaces: Orthogonal Bottlenecks for Reinforcement Learning
May 25, 2026Aleksandar Todorov, Matthia SabatelliRepresentation Learning Reinforcement Learning
University of Groningen, Groningen, The Netherlands
Latent Representation Alignment for Offline Goal-Conditioned Reinforcement Learning
May 25, 2026Hyungkyu Kang, Byeongchan Kim, Min-hwan OhReinforcement Learning Offline RL
Seoul National University, Seoul, South Korea.
Self-supervised Hierarchical Visual Reasoning with World Model
May 17, 2026Yuanfei Xu, Lin Liu, Wengang Zhou +2World Model Learning World Models
Department of Electronic Engineering and Information Science, University of Science and Technology of China, Hefei, Anhui, China
R2R2: Robust Representation for Intensive Experience Reuse via Redundancy Reduction in Self-Predictive Learning
May 13, 2026Sanghyeob Song, Donghyeok Lee, Jinsik Kim +1Reinforcement Learning Self-Supervised Learning
Interdisciplinary Program in Artificial Intelligence, Seoul National University · Samsung Electro-Mechanics, Republic of Korea · Department of Electrical and Computer Engineering, Seoul National University
JEDI: Joint Embedding Diffusion World Model for Online Model-Based Reinforcement Learning
May 13, 2026Jing Yu Lim, Rushi Shah, Zarif Ikram +4World Model Learning Joint-Embedding Predictive Architecture
National University of Singapore · usc
Efficient Hierarchical Implicit Flow Q-learning for Offline Goal-conditioned Reinforcement Learning
Apr 10, 2026Zhiqiang Dong, Teng Pang, Rongjian Xu +1Offline RL Hierarchical RL
School of Software, Shandong University, Jinan, Shandong, China · Luke EI, Jinan, China
Saliency-Guided Representation with Consistency Policy Learning for Visual Unsupervised Reinforcement Learning
Apr 7, 2026Jingbo Sun, Qichao Zhang, Songjun Tu +5Unsupervised RL Reinforcement Learning
SKL-MAIS, Institute of Automation, Chinese Academy of Sciences · Pengcheng Laboratory · School of Artificial Intelligence, University of Chinese A
Stratifying Reinforcement Learning with Signal Temporal Logic
Apr 6, 2026Justin Curry, Alberto SperanzonSignal Temporal Logic Reinforcement Learning
University at Albany, State University of New York, Department of Mathematics and Statistics, 1400 Washington Ave., Albany, NY 12202 USA · Lockheed Martin, Advanced Technology Labs, 1303 Corporate Center Dr, Eagan, MN 55121 USA
Provable Multi-Task Reinforcement Learning: A Representation Learning Framework with Low Rank Rewards
Apr 4, 2026Yaoze Guo, Shana MoothedathMulti-Task RL Representation Learning for RL
Electrical and Computer Engineering, Iowa State University
Automatic feature identification in least-squares policy iteration using the Koopman operator framework
Mar 27, 2026Christian Mugisho Zagabe, Sebastian PeitzRL Control Policy Iteration
Department of Computer Science & Lamarr Institute, TU Dortmund University, Germany
Spectral Alignment in Forward-Backward Representations via Temporal Abstraction
Mar 20, 2026Seyed Mahdi B. Azad, Jasper Hoffmann, Iman Nematollahi +3Reinforcement Learning Representation Learning for RL
Department of Computer Science, University of Freiburg, Germany
R2-Dreamer: Redundancy-Reduced World Models without Decoders or Augmentation
Mar 18, 2026Naoki Morihira, Amal Nahar, Kartik Bharadwaj +3Model-Based RL World Model Learning
Honda R&D Co., Ltd. · The University of Tokyo · RIKEN AIP
Unified Policy Value Decomposition for Rapid Adaptation
Mar 18, 2026Cristiano Capone, Luca Falorsi, Andrea Ciardiello +1Zero-Shot RL Reinforcement Learning
Computational Neuroscience Unit, Istituto Superiore di Sanit`a, 00161, Rome, Italy · Ospedale Santa Lucia, Rome, Italy · School of Computer Science, University of Sheffield,Sheffield,S10 2TN, United Kingdom
Quantum entanglement provides a competitive advantage in adversarial games
Mar 11, 2026Peiyong Wang, Kieran Hymas, James QuachQuantum RL Quantum Machine Learning
Technology, CSIRO Clayton, Research Way, Clayton, VIC 3168, Australia.
Impact of Connectivity on Laplacian Representations in Reinforcement Learning
Mar 9, 2026Tommaso Giorgi, Pierriccardo Olivieri, Keyue Jiang +2Reinforcement Learning Representation Learning for RL
Baker Hughes, Florence, Italy · Politecnico di Milano · University College London, AI Center (UCL) +1
Meta-RL with Shared Representations Enables Fast Adaptation in Energy Systems
Mar 9, 2026Théo Zangato, Aomar Osmani, Pegah AlizadehMeta-Reinforcement Learning Multi-Task RL
Université Sorbonne Paris Nord, France · INSA Rouen, France · Université Lumière Lyon 2, Université Claude Bernard Lyon 1, ERIC, 69007, Lyon, France
DreamSAC: Learning Hamiltonian World Models via Symmetry Exploration
Mar 8, 2026Jinzhou Tang, Fan Feng, Minghao Fu +3World Model Learning World Models
UC San Diego · Sun Yat-sen University
Temporal Consistency Improves Generalization in Contextual Offline Meta Reinforcement Learning
Mar 3, 2026Mohammadreza Nakheai, Aidan Scannell, Kevin Luck +1Representation Learning Offline RL
Aalto University · Toyota Research Institute · Vrije Universiteit Amsterdam

Aug 6, 2026J/K move · Enter open · S save

ProDVI: Programmatic Dynamics Priors for Value Network Initialization

Xinwei Liu, Junyuan Liang, Jianting Zhang, Wuhui Chen

1Sun Yat-sen University · 2Purdue University

Reinforcement Learning LLM-Guided RL Representation Learning for RL Model-Free RL Sample-Efficient RL

Open paper AskPDF arXiv

Deep Reinforcement Learning (RL) is notoriously sample inefficient. One contributing factor is that RL agents are typically initialized from scratch, forcing them to acquire task-relevant knowledge through online interaction. Existing approaches obtain informative initializations through pre-collected datasets, high-fidelity simulators, or meta-learning over related tasks, but these prerequisites may be difficult to access or even unavailable. In this paper, we propose Programmatic Dynamics Priors for Value Network Initialization (ProDVI), a framework that leverages the commonsense and domain knowledge encoded in large language models to initialize RL agents without relying on these resources. Specifically, ProDVI prompts a code-generating language model to produce executable Python functions that encode coarse hypotheses about environment dynamics. These functions are then used to generate synthetic transitions. Based on these transitions, we construct an auxiliary dynamics prediction objective to pretrain the state-action encoder of the value network in an actor-critic framework. The learned representation provides dynamics-aware inductive biases before online RL begins. Importantly, the generated programs are used only for representation pretraining and are not required to faithfully simulate the target environment. While the generated programs may be inaccurate, their induced initialization can be corrected through online learning from real transitions and rewards. Experiments on OpenAI Gym and DeepMind Control Suite tasks show that ProDVI can effectively improve the sample efficiency of model-free RL algorithms.