Abstract: Reinforcement learning (RL) has enabled robots to develop complex skills, but its success in image-based tasks often depends on effective representation learning. Prior works have primarily focused on 2D representations, often overlooking the inherent 3D geometric structure of the world...
Abstract: Skill abstraction, the process of learning reusable and temporally extended behaviors, has emerged as a key focus in robot learning for its potential to improve sample efficiency and generalization. However, existing methods exhibit complementary strengths and weaknesses, typically mode...
Abstract: Multi-task reinforcement learning (MTRL) offers a promising approach to improve sample efficiency and generalization by training agents across multiple tasks, enabling knowledge sharing between them. However, applying MTRL to robotics remains challenging due to the high cost of collecti...