Alert button
Picture for Benjamin Eysenbach

Benjamin Eysenbach

Alert button

Closing the Gap between TD Learning and Supervised Learning -- A Generalisation Point of View

Jan 20, 2024
Raj Ghugare, Matthieu Geist, Glen Berseth, Benjamin Eysenbach

Viaarxiv icon

Bridging State and History Representations: Understanding Self-Predictive RL

Jan 17, 2024
Tianwei Ni, Benjamin Eysenbach, Erfan Seyedsalehi, Michel Ma, Clement Gehring, Aditya Mahajan, Pierre-Luc Bacon

Viaarxiv icon

Contrastive Difference Predictive Coding

Oct 31, 2023
Chongyi Zheng, Ruslan Salakhutdinov, Benjamin Eysenbach

Viaarxiv icon

When Do Transformers Shine in RL? Decoupling Memory from Credit Assignment

Jul 31, 2023
Tianwei Ni, Michel Ma, Benjamin Eysenbach, Pierre-Luc Bacon

Figure 1 for When Do Transformers Shine in RL? Decoupling Memory from Credit Assignment
Figure 2 for When Do Transformers Shine in RL? Decoupling Memory from Credit Assignment
Figure 3 for When Do Transformers Shine in RL? Decoupling Memory from Credit Assignment
Figure 4 for When Do Transformers Shine in RL? Decoupling Memory from Credit Assignment
Viaarxiv icon

Contrastive Example-Based Control

Jul 24, 2023
Kyle Hatch, Benjamin Eysenbach, Rafael Rafailov, Tianhe Yu, Ruslan Salakhutdinov, Sergey Levine, Chelsea Finn

Figure 1 for Contrastive Example-Based Control
Figure 2 for Contrastive Example-Based Control
Figure 3 for Contrastive Example-Based Control
Figure 4 for Contrastive Example-Based Control
Viaarxiv icon

A Connection between One-Step Regularization and Critic Regularization in Reinforcement Learning

Jul 24, 2023
Benjamin Eysenbach, Matthieu Geist, Sergey Levine, Ruslan Salakhutdinov

Viaarxiv icon

HIQL: Offline Goal-Conditioned RL with Latent States as Actions

Jul 22, 2023
Seohong Park, Dibya Ghosh, Benjamin Eysenbach, Sergey Levine

Figure 1 for HIQL: Offline Goal-Conditioned RL with Latent States as Actions
Figure 2 for HIQL: Offline Goal-Conditioned RL with Latent States as Actions
Figure 3 for HIQL: Offline Goal-Conditioned RL with Latent States as Actions
Figure 4 for HIQL: Offline Goal-Conditioned RL with Latent States as Actions
Viaarxiv icon

Stabilizing Contrastive RL: Techniques for Offline Goal Reaching

Jun 06, 2023
Chongyi Zheng, Benjamin Eysenbach, Homer Walke, Patrick Yin, Kuan Fang, Ruslan Salakhutdinov, Sergey Levine

Figure 1 for Stabilizing Contrastive RL: Techniques for Offline Goal Reaching
Figure 2 for Stabilizing Contrastive RL: Techniques for Offline Goal Reaching
Figure 3 for Stabilizing Contrastive RL: Techniques for Offline Goal Reaching
Figure 4 for Stabilizing Contrastive RL: Techniques for Offline Goal Reaching
Viaarxiv icon

Bitrate-Constrained DRO: Beyond Worst Case Robustness To Unknown Group Shifts

Feb 06, 2023
Amrith Setlur, Don Dennis, Benjamin Eysenbach, Aditi Raghunathan, Chelsea Finn, Virginia Smith, Sergey Levine

Figure 1 for Bitrate-Constrained DRO: Beyond Worst Case Robustness To Unknown Group Shifts
Figure 2 for Bitrate-Constrained DRO: Beyond Worst Case Robustness To Unknown Group Shifts
Figure 3 for Bitrate-Constrained DRO: Beyond Worst Case Robustness To Unknown Group Shifts
Figure 4 for Bitrate-Constrained DRO: Beyond Worst Case Robustness To Unknown Group Shifts
Viaarxiv icon