Lifelong learning with weighted majority votes

Pentina, Anastasia and Urner, Ruth (2016) Lifelong learning with weighted majority votes. In: NIPS: Neural Information Processing Systems, December 5-10, 2016, Barcelona, Spain.

[img] Text
main.pdf - Published Version
Available under License All rights reserved.
[IST-2017-775-v1+1]
Download (231Kb)
[img] Text
supplementary.pdf - Supplemental Material
[IST-2017-775-v1+2]
Download (181Kb)
Official URL: http://papers.nips.cc/paper/6095-lifelong-learning...

Abstract

Better understanding of the potential benefits of information transfer and representation learning is an important step towards the goal of building intelligent systems that are able to persist in the world and learn over time. In this work, we consider a setting where the learner encounters a stream of tasks but is able to retain only limited information from each encountered task, such as a learned predictor. In contrast to most previous works analyzing this scenario, we do not make any distributional assumptions on the task generating process. Instead, we formulate a complexity measure that captures the diversity of the observed tasks. We provide a lifelong learning algorithm with error guarantees for every observed task (rather than on average). We show sample complexity reductions in comparison to solving every task in isolation in terms of our task complexity measure. Further, our algorithmic framework can naturally be viewed as learning a representation from encountered tasks with a neural network.

Item Type: Conference or Workshop Item (Paper)
Subjects: 000 Computer science, knowledge & general works > 000 Computer science, knowledge & systems > 006 Special computer methods
Research Group: Lampert Group
SWORD Depositor: Sword Import User
Depositing User: Sword Import User
Date Deposited: 28 Feb 2017 09:05
Last Modified: 05 Sep 2017 09:36
URI: https://repository.ist.ac.at/id/eprint/775

Actions (login required)

View Item View Item