Article

Multitask Learning over Shared Subspaces

2020-07-14

Abstract excerpt

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. We found, as hypothesised, that subject performance was significantly higher on the second task if it shared the same subspace as the first. Additionally, accuracy was positively correlated over subjects lea...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 588073f8-14c3-54d0-8f16-41bc760d9303
DOI: 10.1101/2020.07.12.199265

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Multitask Learning over Shared SubspacesDOI 10.1101/2020.07.12.199265Read

Humans and neural networks show similar patterns of transfer and interference during continual learningSemantic proximity Read

Decoding task representations that support generalization in hierarchical taskSemantic proximity Read

Identifying Transfer Learning in the Reshaping of Inductive BiasesSemantic proximity Read

System alignment supports cross-domain learning and zero-shot generalisationSemantic proximity Read

Cracking the multidimensionality of cognition with cross-domain transferSemantic proximity Read

Multi-Task Reinforcement Learning in HumansSemantic proximity Read

Beyond independent latent classes: Testing the limits of human flexibilitySemantic proximity Read

Decomposing dynamical subprocesses for compositional generalizationSemantic proximity Read

Generalizing to generalize: when (and when not) to be compositional in task structure learningSemantic proximity Read

Similarities and differences in spatial and non-spatial cognitive mapsSemantic proximity Read

Compositional clustering in task structure learningSemantic proximity Read

A mixture of generative models strategy helps humans generalize across tasksSemantic proximity Read