Pieter Abbeel
- A Stable Whitening Optimizer for Efficient Neural Network Training
- Bigger, Regularized, Categorical: High-Capacity Value Functions are Efficient Multi-Task Learners
- Coarse-to-fine Q-Network with Action Sequence for Data-Efficient Reinforcement Learning
- Compute-Optimal Scaling for Value-Based Deep RL
- DexGarmentLab: Dexterous Garment Manipulation Environment with Generalizable Policy
- Object-centric 3D Motion Field for Robot Learning from Human Videos
- SOMBRL: Scalable and Optimistic Model-Based RL