Lifelong Learning using Eigentasks: Task Separation, Skill Acquisition and Selective Transfer

Aswin Raghavan, Jesse Hostetler, Indranil Sur, Abrar Rahman, Ajay Divakaran

Student First Author: No

Keywords: Lifelong Learning, Reinforcement Learning, Continual Learning, Machine Learning, Starcraft

Abstract: We introduce the eigentask framework for lifelong learning. An eigentask is a pairing of a skill that solves a set of related tasks with a generative model that can sample from the skill's input space. The model extends generative replay approaches, which have mainly been used to avoid catastrophic forgetting, to also address other lifelong learning goals such as forward knowledge transfer. We propose a wake-sleep cycle of alternating task learning and knowledge consolidation for learning in our framework, and instantiate it for lifelong supervised learning and lifelong RL. We achieve performance comparable to the state-of-the-art in supervised continual learning benchmarks, and show encouraging preliminary results for forward knowledge transfer in a lifelong RL application in the video game Starcraft~2.

TL;DR: Generative model + Mixture of experts = Eigentask