Bayesian Active Learning for Maximal Information Gain on Model Parameters

Kasra Arnavaz, Aasa Feragen, Oswin Krause, Marco Loog

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

7 Downloads (Pure)

Abstract

The fact that machine learning models, despite their advancements, are still trained on randomly gathered data is proof that a lasting solution to the problem of optimal data gathering has not yet been found. In this paper, we investigate whether a Bayesian approach to the classification problem can provide assumptions under which one is guaranteed to perform at least as good as random sampling. For a logistic regression model, we show that maximal expected information gain on model parameters is a promising criterion for selecting samples, assuming that our classification model is well-matched to the data. Our derived criterion is closely related to the maximum model change. We experiment with data sets which satisfy this assumption to varying degrees to see how sensitive our performance is to the violation of our assumption in practice.
Original languageEnglish
Title of host publicationProceedings of 2020 25th International Conference on Pattern Recognition
PublisherIEEE
Publication date2021
Pages10524-10531
ISBN (Print)9781728188089
DOIs
Publication statusPublished - 2021
Event25th International Conference on Pattern Recognition - Virtual event, Milano, Italy
Duration: 10 Jan 202115 Jan 2021

Conference

Conference25th International Conference on Pattern Recognition
LocationVirtual event
Country/TerritoryItaly
CityMilano
Period10/01/202115/01/2021
Series2020 25th International Conference on Pattern Recognition (icpr)

Fingerprint

Dive into the research topics of 'Bayesian Active Learning for Maximal Information Gain on Model Parameters'. Together they form a unique fingerprint.

Cite this