Defense: Learning Hierarchical Abstractions from Human Demonstrations for Application-Scale Domains

Speaker Name: 
Michael Leece
Speaker Title: 
Ph.D. Candidate (Advisor: Arnav Jhala)
Speaker Organization: 
Computer Science and Engineering Department
Start Time: 
Monday, September 10, 2018 - 7:00am
Arnav Jhala

As the collection of data becomes more and more commonplace, it unlocks new approaches to old problems in the field of artificial intelligence. Much of this benefit is realized by advances in the decision problems of machine learning and statistics, but value can be gleaned for more classical AI problems as well. Large databases of human demonstrations allow for bootstrapping planning models in complex domains that previously would have been computationally infeasible.
In this talk, I will present algorithms and systems for learning planning abstractions from human demonstrations in real-time-strategy (RTS) games. These are intricate domains that require reasoning at multiple levels of abstraction, and have been deemed one of the next challenge domains for artificial agents following the success of DeepMind's AlphaGo. The particular challenges that will arise in this work are primarily due to the inconsistency of human planning and variations in style and skill level between human players, and also the complexity of the domain. Any algorithm that intends to learn from human data must overcome these hurdles. The presented approaches draw inspiration from a number of machine learning algorithms and paradigms, which have been developed explicitly in the context of large-scale, noisy data.