In plain words: A broad review sorts through recent algorithms that learn by copying an expert's actions, laying out how each one works. It compares their performance and how much worse they do than the expert over time, rather than just listing them.

Abstract

Imitation Learning is a sequential task where the learner tries to mimic an expert's action in order to achieve the best performance. Several algorithms have been proposed recently for this task. In this project, we aim at proposing a wide review of these algorithms, presenting their main features and comparing them on their performance and their regret bounds.

Alexandre Attia, Sharone Dayan
arXiv:1801.06503 · stat.ML, cs.LG · submitted Jan 19, 2018
abstract · pdf · html · 9 pages, 5 figures, 5 appendix pages