UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Incremental learning-to-learn with statistical guarantees

Denevi, G; Ciliberto, C; Stamos, D; Pontil, M; (2018) Incremental learning-to-learn with statistical guarantees. In: Globerson, Amir and Silva, Ricardo, (eds.) Proceedings of the Thirty-Fourth Conference (2018), Uncertainty in Artificial Intelligence. (pp. pp. 457-466). AUAI: California, USA. Green open access

[thumbnail of 181.pdf]
Preview
Text
181.pdf - Accepted version

Download (1MB) | Preview

Abstract

In learning-to-learn the goal is to infer a learning algorithm that works well on a class of tasks sampled from an unknown metadistribution. In contrast to previous work on batch learning-to-learn, we consider a scenario where tasks are presented sequentially and the algorithm needs to adapt incrementally to improve its performance on future tasks. Key to this setting is for the algorithm to rapidly incorporate new observations into the model as they arrive, without keeping them in memory. We focus on the case where the underlying algorithm is Ridge Regression parametrised by a symmetric positive semidefinite matrix. We propose to learn this matrix by applying a stochastic strategy to minimize the empirical error incurred by Ridge Regression on future tasks sampled from the meta-distribution. We study the statistical properties of the proposed algorithm and prove non-asymptotic bounds on its excess transfer risk, that is, the generalization performance on new tasks from the same meta-distribution. We compare our online learning-to-learn approach with a state-of-the-art batch method, both theoretically and empirically.

Type: Proceedings paper
Title: Incremental learning-to-learn with statistical guarantees
Event: Thirty-Fourth Conference (2018), Uncertainty in Artificial Intelligence
ISBN-13: 9781510871601
Open access status: An open access version is available from UCL Discovery
Publisher version: http://auai.org/uai2018/proceedings/uai2018proceed...
Additional information: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science
URI: https://discovery.ucl.ac.uk/id/eprint/10066090
Downloads since deposit
26Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item