Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Imitation Bootstrapped Reinforcement Learning

About

Despite the considerable potential of reinforcement learning (RL), robotic control tasks predominantly rely on imitation learning (IL) due to its better sample efficiency. However, it is costly to collect comprehensive expert demonstrations that enable IL to generalize to all possible scenarios, and any distribution shift would require recollecting data for finetuning. Therefore, RL is appealing if it can build upon IL as an efficient autonomous self-improvement procedure. We propose imitation bootstrapped reinforcement learning (IBRL), a novel framework for sample-efficient RL with demonstrations that first trains an IL policy on the provided demonstrations and then uses it to propose alternative actions for both online exploration and bootstrapping target values. Compared to prior works that oversample the demonstrations or regularize RL with an additional imitation loss, IBRL is able to utilize high quality actions from IL policies since the beginning of training, which greatly accelerates exploration and training efficiency. We evaluate IBRL on 6 simulation and 3 real-world tasks spanning various difficulty levels. IBRL significantly outperforms prior methods and the improvement is particularly more prominent in harder tasks.

Hengyuan Hu, Suvir Mirchandani, Dorsa Sadigh• 2023

Related benchmarks

TaskDatasetResultRank
LiftRobomimic Lift-State
Success Rate100
30
Square Nut AssemblyRobomimic Square-State
Success Rate94
30
Can Pick & PlaceRobomimic Can-State
Success Rate54
30
Dexterous ManipulationAdroit Pen
Success Rate95
26
Robotic ManipulationCan-Image
Success Rate12
21
Kitchen manipulationD4RL kitchen
Success Rate50
18
Door manipulationD4RL Door
Success Rate0.00e+0
18
Robotic ManipulationLift-Image
Success Rate100
14
Robotic Manipulationram 60% illumination shift
Success Rate (SR)43
12
Dexterous ManipulationDexterous Manipulation Simulation (test)
Grasping40.4
12
Showing 10 of 13 rows

Other info

Follow for update