We would like to congratulate 2021 ACM Prize in Computing Recipient Pieter Abbeel, who was recognized "for contributions to robot learning, including learning from demonstrations and deep reinforcement learning for robotic control." Abbeel’s work pioneered teaching robots to learn from human demonstrations ("apprenticeship learning") and through their own trial and error ("reinforcement learning"). His contributions include developing robots capable of performing surgical suturing, detecting objects, and planning their trajectories in uncertain situations. In addition, he has pioneered "few-shot imitation learning," where a robot is able to learn a task from just one demonstration after pre-training with a large set of demonstrations on related tasks.

more