Pieter Abbeel

Behavioral Priors and Dynamics Models: Improving Performance and Domain Transfer in Offline RL

Jun 18, 2021
Catherine Cang, Aravind Rajeswaran, Pieter Abbeel, Michael Laskin

Figure 2 for Behavioral Priors and Dynamics Models: Improving Performance and Domain Transfer in Offline RL

Figure 3 for Behavioral Priors and Dynamics Models: Improving Performance and Domain Transfer in Offline RL

Figure 4 for Behavioral Priors and Dynamics Models: Improving Performance and Domain Transfer in Offline RL

Access Paper or Ask Questions

Unsupervised Learning of Visual 3D Keypoints for Control

Jun 14, 2021
Boyuan Chen, Pieter Abbeel, Deepak Pathak

Figure 2 for Unsupervised Learning of Visual 3D Keypoints for Control

Figure 3 for Unsupervised Learning of Visual 3D Keypoints for Control

Figure 4 for Unsupervised Learning of Visual 3D Keypoints for Control

* Accepted at ICML 2021. Videos and code at https://buoyancy99.github.io/unsup-3d-keypoints/

Access Paper or Ask Questions

PEBBLE: Feedback-Efficient Interactive Reinforcement Learning via Relabeling Experience and Unsupervised Pre-training

Jun 09, 2021
Kimin Lee, Laura Smith, Pieter Abbeel

Figure 2 for PEBBLE: Feedback-Efficient Interactive Reinforcement Learning via Relabeling Experience and Unsupervised Pre-training

Figure 3 for PEBBLE: Feedback-Efficient Interactive Reinforcement Learning via Relabeling Experience and Unsupervised Pre-training

Figure 4 for PEBBLE: Feedback-Efficient Interactive Reinforcement Learning via Relabeling Experience and Unsupervised Pre-training

* ICML 2021. First two authors contributed equally. Website: https://sites.google.com/view/icml21pebble Code: https://github.com/pokaxpoka/B_Pref

Access Paper or Ask Questions

Decision Transformer: Reinforcement Learning via Sequence Modeling

Jun 02, 2021
Lili Chen, Kevin Lu, Aravind Rajeswaran, Kimin Lee, Aditya Grover, Michael Laskin, Pieter Abbeel, Aravind Srinivas, Igor Mordatch

Figure 2 for Decision Transformer: Reinforcement Learning via Sequence Modeling

Figure 3 for Decision Transformer: Reinforcement Learning via Sequence Modeling

Figure 4 for Decision Transformer: Reinforcement Learning via Sequence Modeling

* First two authors contributed equally. Last two authors advised equally

Access Paper or Ask Questions

JUMBO: Scalable Multi-task Bayesian Optimization using Offline Data

Jun 02, 2021
Kourosh Hakhamaneshi, Pieter Abbeel, Vladimir Stojanovic, Aditya Grover

Figure 2 for JUMBO: Scalable Multi-task Bayesian Optimization using Offline Data

Figure 3 for JUMBO: Scalable Multi-task Bayesian Optimization using Offline Data

Figure 4 for JUMBO: Scalable Multi-task Bayesian Optimization using Offline Data

Access Paper or Ask Questions

Learning What To Do by Simulating the Past

May 03, 2021
David Lindner, Rohin Shah, Pieter Abbeel, Anca Dragan

Figure 2 for Learning What To Do by Simulating the Past

Figure 3 for Learning What To Do by Simulating the Past

Figure 4 for Learning What To Do by Simulating the Past

* Presented at ICLR 2021

Access Paper or Ask Questions

VideoGPT: Video Generation using VQ-VAE and Transformers

Apr 20, 2021
Wilson Yan, Yunzhi Zhang, Pieter Abbeel, Aravind Srinivas

Figure 2 for VideoGPT: Video Generation using VQ-VAE and Transformers

Figure 3 for VideoGPT: Video Generation using VQ-VAE and Transformers

Figure 4 for VideoGPT: Video Generation using VQ-VAE and Transformers

* Project website: https://wilson1yan.github.io/videogpt/index.html

Access Paper or Ask Questions

Auto-Tuned Sim-to-Real Transfer

Apr 15, 2021
Yuqing Du, Olivia Watkins, Trevor Darrell, Pieter Abbeel, Deepak Pathak

Figure 2 for Auto-Tuned Sim-to-Real Transfer

Figure 3 for Auto-Tuned Sim-to-Real Transfer

Figure 4 for Auto-Tuned Sim-to-Real Transfer

* ICRA 2021. First two authors contributed equally. Website at https://yuqingd.github.io/autotuned-sim2real/

Access Paper or Ask Questions

GEM: Group Enhanced Model for Learning Dynamical Control Systems

Apr 07, 2021
Philippe Hansen-Estruch, Wenling Shang, Lerrel Pinto, Pieter Abbeel, Stas Tiomkin

Figure 2 for GEM: Group Enhanced Model for Learning Dynamical Control Systems

Figure 3 for GEM: Group Enhanced Model for Learning Dynamical Control Systems

Figure 4 for GEM: Group Enhanced Model for Learning Dynamical Control Systems

* 14 pages, 8 figures

Access Paper or Ask Questions