DATA SCIENCE & SYSTEM SECURITY
PROJECTS
PEOPLE
PUBLICATIONS
PATENTS
This project aims to learn skills by mimicking experts’ behaviors in given tasks. The proposed SAiL engine is trained to perform action prediction tasks from demonstrations by learning a mapping function between observed states and actions. The main challenges in real applications, medical and health care, for example, are that the collection of such experts’ demonstrations is very expensive, and it takes a large amount of time and money for expert training.
Additionally, the demonstration varies among experts, making it hard to reuse and interpret. Each individual expert’s decision-making process can also be affected by different factors and may not be consistent or easy to explain.
Our SAiL project provides the solutions to the aforementioned issues from the following critical dimensions.
Team Members: Wenchao Yu, Haifeng Chen
Publication Tags: skill discovery, SAiL
Imitation learning has achieved great success in many sequential decision-making tasks, in which a neural agent is learned by imitating collected human demonstrations. However, existing algorithms typically require a large number of high-quality demonstrations that are difficult and expensive to collect.
