Shapley Based Residual Decomposition for Instance Analysis

dc.contributor.authorLiu, Tommyen
dc.contributor.authorBarnard, Amandaen
dc.date.accessioned2025-06-24T01:36:49Z
dc.date.available2025-06-24T01:36:49Z
dc.date.issued2023en
dc.description.abstractIn this paper, we introduce the idea of decomposing the residuals of regression with respect to the data instances instead of features. This allows us to determine the effects of each individual instance on the model and each other, and in doing so makes for a model-agnostic method of identifying instances of interest. In doing so, we can also determine the appropriateness of the model and data in the wider context of a given study. The paper focuses on the possible applications that such a framework brings to the relatively unexplored field of instance analysis in the context of Explainable AI tasks.en
dc.description.sponsorshipThis research/project was undertaken with the assistance of resources and services from the National Computational Infrastructure (NCI), which is supported by the Australian Government, along with support by an Australian Government Research Training Program (RTP) Scholarship.en
dc.description.statusPeer-revieweden
dc.format.extent22en
dc.identifier.otherORCID:/0000-0002-4784-2382/work/161120551en
dc.identifier.scopus85174410080en
dc.identifier.urihttp://www.scopus.com/inward/record.url?scp=85174410080&partnerID=8YFLogxKen
dc.identifier.urihttps://hdl.handle.net/1885/733764611
dc.language.isoenen
dc.relation.ispartofseries40th International Conference on Machine Learning, ICML 2023en
dc.rightsPublisher Copyright: © 2023 Proceedings of Machine Learning Research. All rights reserved.en
dc.sourceProceedings of Machine Learning Researchen
dc.titleShapley Based Residual Decomposition for Instance Analysisen
dc.typeConference paperen
dspace.entity.typePublicationen
local.bibliographicCitation.lastpage21936en
local.bibliographicCitation.startpage21915en
local.contributor.affiliationLiu, Tommy; School of Computing, ANU College of Systems and Society, The Australian National Universityen
local.contributor.affiliationBarnard, Amanda; School of Computing, ANU College of Systems and Society, The Australian National Universityen
local.identifier.ariespublicationa383154xPUB44236en
local.identifier.citationvolume202en
local.identifier.pure772f03ca-6db7-4a1e-8d22-6f7b86b2077een
local.identifier.urlhttps://www.scopus.com/pages/publications/85174410080en
local.type.statusPublisheden

Downloads