search menu icon-carat-right cmu-wordmark

A Series of Unlikely Events: Learning Patterns by Observing Sequential Behavior (video)

November 2019 Video
Eric Heim

A Series of Unlikely Events: Learning Patterns by Observing Sequential Behavior

Publisher:

Software Engineering Institute

Watch

Abstract

In this work, We propose an alternative approach, inverse reinforcement learning (IRL), that observes all states and actions in data and computes a statistical model of the world that includes whether each behavior is part of a routine. Deviations from routines have a low likelihood of
occurrence within the model.