BE
Benjamin Eysenbach
cs.LGcs.AIcs.ROstat.MLcs.CVcond-mat.stat-mechcs.HCphysics.bio-phcond-mat.dis-nncond-mat.soft
On Valency
published · living versionsW_4ur66ypq·v1 · currentpublished
Geometric Time Score Estimators: Bridging Self-Predictive Representations and Generative Occupancy Models For Reinforcement Learning
with Aravind Venugopal, Grace Liu, Jeff Schneider
1 version
W_xkdnmba4·v1published
W_vry4nk79·v1published
Preprints & journals
91 papers in the corpus · 2016–2026Training Controllable World Models via Action Influence Maximization2610.27387v1 · Samuel M. Bateman, Tenny Yin, Chongyi Zheng et al.2026 · 0 citationsarXivon Valency
Measuring Empowerment from Offline Data for Intrinsically Motivated Exploration2610.24461v1 · Ishir Garg, Vivek Myers, Qiyang Li et al.2026 · 0 citationsarXivon Valency
Hyperspherical Contrastive Representations Unlock Planning and Exploration2610.20409v1 · Jonghae Park, Sungil Son, Jonghun Shin et al.2026 · 0 citationsarXiv
Scaling Empowered Agents2610.13002v1 · Catherine Ji, Vivek Myers, Joshua Lin et al.2026 · 0 citationsarXiv
A Unified Bellman Operator for Safety-Critical Reinforcement Learning2610.12420v1 · Nishanth Arun Rao, Royina Karegoudra Jayanth, Benjamin Eysenbach et al.2026 · 0 citationsarXiv
The Geometry of Empowerment2610.07796v2 · Catherine Ji, Vivek Myers, Sergey Levine et al.2026 · 0 citationsarXiv
Up and Down the Abstraction Ladder: Code-Based Skills for Language Agents2609.31076v1 · Bart\lomiej Cupiał, Jens Tuyls, Maciej Wo\lczyk et al.2026 · 0 citationsarXiv
Bridging Control, Inference, Transport, and Thermodynamics: From Theory to Applications in Learning.42780301 · Blumenthal, Emmy, Claussen, Nikolas, Eysenbach, Benjamin et al.2026 · 0 citationsArXiv. 2026
Three Steps at a Time: Learning Representations from Action Sequences in Contrastive RL2608.30640v1 · Michal Korniak, Kamil Dybek, Benjamin Eysenbach et al.2026 · 0 citationsarXiv
LeAct: Learning to Reason from Expert Actions2607.21856v1 · Ziran Yang, Chengshuai Shi, Raj Ghugare et al.2026 · 0 citationsarXiv
Learning to Perceive the World Through Control: Empowerment-Based Representation Learning2605.30656v1 · Mahsa Bastankhah, Sophie Broderick, Benjamin Eysenbach2026 · 0 citationsarXiv
Behavior-Consistent Deep Reinforcement Learning2605.21214v3 · Marcel Hussing, Liv G. d'Aliberti, Claas Voelcker et al.2026 · 0 citationsarXiv
Unifying Goal-Conditioned RL and Unsupervised Skill Learning via Control-Maximization2605.06145v2 · Alireza Modirshanechi, Benjamin Eysenbach, Peter Dayan et al.2026 · 0 citationsarXiv
Occupancy Reward Shaping: Improving Credit Assignment for Offline Goal-Conditioned Reinforcement Learning2604.20627v1 · Aravind Venugopal, Jiayu Chen, Xudong Wu et al.2026 · 0 citationsarXiv
Temporal Representations for Exploration: Learning Complex Exploratory Behavior without Extrinsic Rewards2603.02008v2 · Faisal Mohamed, Catherine Ji, Benjamin Eysenbach et al.2026 · 0 citationsarXiv
UpSkill: Mutual Information Skill Learning for Structured Response Diversity in LLMs2602.22296v1 · Devan Shah, Owen Yang, Daniel Yang et al.2026 · 0 citationsarXiv
Can We Really Learn One Representation to Optimize All Rewards?2602.11399v2 · Chongyi Zheng, Royina Karegoudra Jayanth, Benjamin Eysenbach2026 · 0 citationsarXiv
On the Role of Computation in Reinforcement Learning2602.05999v4 · Raj Ghugare, MichałBortkiewicz, Alicja Ziarko et al.2026 · 0 citationsarXiv
Multistep Quasimetric Learning for Scalable Goal-conditioned Reinforcement Learning2511.07730v3 · Bill Chunyuan Zheng, Vivek Myers, Benjamin Eysenbach et al.2025 · 0 citationsICLR (2026)
Is Temporal Difference Learning the Gold Standard for Stitching in RL?2510.21995v1 · MichałBortkiewicz, W\ladys\law Pa\lucki, Mateusz Ostaszewski et al.2025 · 0 citationsarXiv
Consistent Zero-Shot Imitation with Contrastive Goal Inference2510.17059v2 · Kathryn Wantlin, Chongyi Zheng, Benjamin Eysenbach2025 · 0 citationsarXiv
Training LLM Agents to Empower Humans2510.13709v2 · Evan Ellis, Vivek Myers, Jens Tuyls et al.2025 · 0 citationsarXiv
Demystifying the Mechanisms Behind Emergent Exploration in Goal-conditioned RL2510.14129v1 · Mahsa Bastankhah, Grace Liu, Dilip Arumugam et al.2025 · 0 citationsarXiv
Value Flows2510.07650v4 · Perry Dong, Chongyi Zheng, Chelsea Finn et al.2025 · 0 citationsarXiv
BuilderBench: The Building Blocks of Intelligent Agents2510.06288v4 · Raj Ghugare, Roger Creus Castanyer, Catherine Ji et al.2025 · 0 citationsarXiv
Offline Goal-conditioned Reinforcement Learning with Quasimetric Representations2509.20478v2 · Vivek Myers, Bill Chunyuan Zheng, Benjamin Eysenbach et al.2025 · 2 citationsarXiv
Self-Supervised Goal-Reaching Results in Multi-Agent Cooperation and Exploration2509.10656v2 · Chirayu Nimonkar, Shlok Shah, Catherine Ji et al.2025 · 0 citationsarXiv
Contrastive Representations for Temporal Reasoning2508.13113v2 · Alicja Ziarko, Michal Bortkiewicz, Michal Zawalski et al.2025 · 0 citationsarXiv
Identifying nonequilibrium degrees of freedom in high-dimensional stochastic systems2508.08247v2 · Catherine Ji, Ravin Raj, Benjamin Eysenbach et al.2025 · 0 citationsarXiv
Skill Learning via Policy Diversity Yields Identifiable Representations for Reinforcement Learning2507.14748v1 · Patrik Reizinger, B'alint Mucs'anyi, Siyuan Guo et al.2025 · 0 citationsarXiv
Intention-Conditioned Flow Occupancy Models2506.08902v4 · Chongyi Zheng, Seohong Park, Sergey Levine et al.2025 · 0 citationsarXiv
Horizon Reduction Makes RL Scalable2506.04168v3 · Seohong Park, Kevin Frans, Deepinder Mann et al.2025 · 0 citationsarXiv
Normalizing Flows are Capable Models for Continuous Control2505.23527v4 · Raj Ghugare, Benjamin Eysenbach2025 · 0 citationsarXiv
1000 Layer Networks for Self-Supervised RL: Scaling Depth Can Enable New Goal-Reaching Capabilities2503.14858v5 · Kevin Wang, Ishaan Javali, MichałBortkiewicz et al.2025 · 0 citationsarXiv
The "Law" of the Unconscious Contrastive Learner: Probabilistic Alignment of Unpaired Modalities2501.11326v1 · Yongwei Che, Benjamin Eysenbach2025 · 0 citationsarXiv
Horizon Generalization in Reinforcement Learning2501.02709v1 · Vivek Myers, Catherine Ji, Benjamin Eysenbach2025 · 0 citationsInternational Conference on Learning Representations (ICLR), 2025
Can a MISL Fly? Analysis and Ingredients for Mutual Information Skill Learning2412.08021v3 · Chongyi Zheng, Jens Tuyls, Joanne Peng et al.2024 · 0 citationsarXiv
Learning to Assist Humans without Inferring Rewards2411.02623v3 · Vivek Myers, Evan Ellis, Sergey Levine et al.2024 · 0 citationsarXiv
GHIL-Glue: Hierarchical Control with Filtered Subgoal Images2410.20018v1 · Kyle B. Hatch, Ashwin Balakrishna, Oier Mees et al.2024 · 2 citationsarXiv
OGBench: Benchmarking Offline Goal-Conditioned RL2410.20092v2 · Seohong Park, Kevin Frans, Benjamin Eysenbach et al.2024 · 0 citationsarXiv
Accelerating Goal-Conditioned RL Algorithms and Research2408.11052v4 · MichałBortkiewicz, W\ladys\law Pa\lucki, Vivek Myers et al.2024 · 0 citationsInternational Conference on Learning Representations (ICLR), 2025
A Single Goal is All You Need: Skills and Exploration Emerge from Contrastive RL without Rewards, Demonstrations, or Subgoals2408.05804v1 · Grace Liu, Michael Tang, Benjamin Eysenbach2024 · 0 citationsarXiv
Learning Temporal Distances: Contrastive Successor Features Can Provide a Metric Structure for Decision-Making2406.17098v2 · Vivek Myers, Chongyi Zheng, Anca Dragan et al.2024 · 0 citationsProceedings of the 41st International Conference on Machine Learning (ICML 2024)
A Rate-Distortion View of Uncertainty Quantification2406.10775v2 · Ifigeneia Apostolopoulou, Benjamin Eysenbach, Frank Nielsen et al.2024 · 1 citationInternational Conference on Machine Learning, 2024
Inference via Interpolation: Contrastive Representations Provably Enable Planning and Inference2403.04082v4 · Benjamin Eysenbach, Vivek Myers, Ruslan Salakhutdinov et al.2024 · 1 citationNeural information processing systems (2024)
Closing the Gap between TD Learning and Supervised Learning -- A Generalisation Point of View2401.11237v2 · Raj Ghugare, Matthieu Geist, Glen Berseth et al.2024 · 0 citationsarXiv
Bridging State and History Representations: Understanding Self-Predictive RL2401.08898v3 · Tianwei Ni, Benjamin Eysenbach, Erfan Seyedsalehi et al.2024 · 3 citationsarXiv
Contrastive Difference Predictive Coding2310.20141v3 · Chongyi Zheng, Ruslan Salakhutdinov, Benjamin Eysenbach2023 · 1 citationarXiv
A Connection between One-Step Regularization and Critic Regularization in Reinforcement Learning2307.12968v1 · Benjamin Eysenbach, Matthieu Geist, Sergey Levine et al.2023 · 1 citationarXiv
Contrastive Example-Based Control2307.13101v1 · Kyle Hatch, Benjamin Eysenbach, Rafael Rafailov et al.2023 · 0 citationsProceedings of The 5th Annual Learning for Dynamics and Control Conference, PMLR 211:155-169, 2023
Career total: 101 works. 91 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.