Bridging Lifelong and Multi-Task Representation Learning via Algorithm and Complexity Measure

TL;DR

This paper introduces a new framework and algorithm for lifelong representation learning, leveraging a novel complexity measure called task-eluder dimension to analyze sample efficiency across various learning problems.

Contribution

It proposes a simple algorithm for lifelong learning that uses multi-task empirical risk minimization and introduces the task-eluder dimension for complexity analysis.

Findings

Sample complexity bounds depend on task-eluder dimension.

Applicable to classification and regression with noise.

Framework unifies lifelong and multi-task learning insights.

Abstract

In lifelong learning, a learner faces a sequence of tasks with shared structure and aims to identify and leverage it to accelerate learning. We study the setting where such structure is captured by a common representation of data. Unlike multi-task learning or learning-to-learn, where tasks are available upfront to learn the representation, lifelong learning requires the learner to make use of its existing knowledge while continually gathering partial information in an online fashion. In this paper, we consider a generalized framework of lifelong representation learning. We propose a simple algorithm that uses multi-task empirical risk minimization as a subroutine and establish a sample complexity bound based on a new notion we introduce--the task-eluder dimension. Our result applies to a wide range of learning problems involving general function classes. As concrete examples, we instantiate our result on classification and regression tasks under noise.