Multitask learning over shared subspaces

Menghi, Nicholas, Kacar, Kemal and Penny, Will (2021) Multitask learning over shared subspaces. PLoS Computational Biology, 17 (7). ISSN 1553-734X

[img]
Preview
PDF (Published_Version) - Published Version
Available under License Creative Commons Attribution.

Download (1MB) | Preview

Abstract

This paper uses constructs from machine learning to define pairs of learning tasks that either shared or did not share a common subspace. Human subjects then learnt these tasks using a feedback-based approach and we hypothesised that learning would be boosted for shared subspaces. Our findings broadly supported this hypothesis with either better performance on the second task if it shared the same subspace as the first, or positive correlations over task performance for shared subspaces. These empirical findings were compared to the behaviour of a Neural Network model trained using sequential Bayesian learning and human performance was found to be consistent with a minimal capacity variant of this model. Networks with an increased representational capacity, and networks without Bayesian learning, did not show these transfer effects. We propose that the concept of shared subspaces provides a useful framework for the experimental study of human multitask and transfer learning.

Item Type: Article
Uncontrolled Keywords: ecology, evolution, behavior and systematics,ecology,modelling and simulation,molecular biology,genetics,cellular and molecular neuroscience,computational theory and mathematics ,/dk/atira/pure/subjectarea/asjc/1100/1105
Faculty \ School: Faculty of Social Sciences > School of Psychology
Related URLs:
Depositing User: LivePure Connector
Date Deposited: 22 Jul 2021 00:09
Last Modified: 30 Sep 2021 16:41
URI: https://ueaeprints.uea.ac.uk/id/eprint/80704
DOI: 10.1371/journal.pcbi.1009092

Actions (login required)

View Item View Item