Grounded action transformation for sim-to-real reinforcement learning

Grounded Action Transformation for Robot Learning in Simulation

Robot learning in simulation is a promising alternative to the prohibitive sample cost of learning in the physical world. Unfortunately, policies learned in simulation often perform worse than hand-coded policies when applied on the physical robot. This paper proposes a new algorithm for learning in simulation – Grounded Action Transformation – and applies it to learning of humanoid bipedal loc...

Grounded Action Transformation for Robot Learning in Simulation

Robot learning in simulation is a promising alternative to the prohibitive sample cost of learning in the physical world. Unfortunately, policies learned in simulation often perform worse than hand-coded policies when applied on the physical robot. Grounded simulation learning (GSL) promises to address this issue by altering the simulator to better match the real world. This paper proposes a ne...

Flexible Robotic Grasping with Sim-to-Real Transfer based Reinforcement Learning

Robotic manipulation requires a highly flexible and compliant system. Task-specific heuristics are usually not able to cope with the diversity of the world outside of specific assembly lines and cannot generalize well. Reinforcement learning methods provide a way to cope with uncertainty and allow robots to explore their action space to solve specific tasks. However, this comes at a cost of hig...

Learning Behaviorally Grounded State Representations for Reinforcement Learning Agents

The learning and reasoning capabilities of biological systems by far exceed those of robots and artificial agents. Part of this stems from their ability to efficiently learn behavioral skills and increasingly complex, symbolic representations that capture the important aspects of their environment. This paper presents an autonomous learning approach by which artificial reinforcement learning ag...

Action discovery for reinforcement learning