April 2026
Robotic World Model: Toward Real-World Online Reinforcement Learning
Toward real-world online reinforcement learning using model-based methods.
Recent & Upcoming Talks
January 2025
Developmental Legged Intelligence: from Imitation to Generation
Popular methods that enable robot learning from limited demonstrations.
January 2025
Developmental Legged Intelligence: from Imitation to Generation
Popular methods that enable robot learning from limited demonstrations.
March 2024
FLD: Fourier Latent Dynamics for Structured Motion Representation and Learning
A self-supervised, structured representation and generation method that extracts spatial-temporal relationships in periodic or quasi-periodic motions.
March 2024
Journal Club on Pose Estimation and Action Recognition
A self-supervised, structured representation and generation method that extracts spatial-temporal relationships in periodic or quasi-periodic motions.
March 2024
Leveraging Spatial-Temporal Structures in Self-supervised Motion Representation and Learning
A self-supervised, structured representation and generation method that extracts spatial-temporal relationships in periodic or quasi-periodic motions.
March 2024
Developmental Legged Intelligence: from Imitation to Generation
Popular methods that enable robot learning from limited demonstrations.
November 2023
Learning from Limited Demonstrations: Lessons We Have Learned
Popular methods that enable robot learning from limited demonstrations.
January 2023
Reducing Human Efforts in Reinforcement Learning for Legged Skill Development
Works that propose possible solutions to reduce human involvement, specifically with examples of legged skill development using model-free reinforcement learning and imitation learning methods.
December 2022
Reducing Human Efforts for Expert Acquisition in Imitation Learning
Methods that allow skill transfer from limited and noisy reference demonstrations.
August 2022
Learning Agile Skills via Adversarial Imitation of Rough Partial Demonstrations
A method for inferring imitation signals from poor expert demonstrations.