Active Learning for Decision-Making from Imbalanced Observational Data

 |  Login

Show simple item record

dc.contributor Aalto-yliopisto fi
dc.contributor Aalto University en
dc.contributor.author Sundin, Iiris
dc.contributor.author Schulam, Peter
dc.contributor.author Siivola, Eero
dc.contributor.author Vehtari, Aki
dc.contributor.author Saria, Suchi
dc.contributor.author Kaski, Samuel
dc.date.accessioned 2019-07-30T07:18:56Z
dc.date.available 2019-07-30T07:18:56Z
dc.date.issued 2019
dc.identifier.citation Sundin , I , Schulam , P , Siivola , E , Vehtari , A , Saria , S & Kaski , S 2019 , Active Learning for Decision-Making from Imbalanced Observational Data . in Proceedings of the 36th International Conference on Machine Learning . Proceedings of Machine Learning Research , vol. 97 , PMLR , International Conference on Machine Learning , Long Beach , United States , 10/06/2019 . en
dc.identifier.issn 1938-7228
dc.identifier.other PURE UUID: a315d279-4aac-4e0a-a5bf-649665848da1
dc.identifier.other PURE ITEMURL: https://research.aalto.fi/en/publications/active-learning-for-decisionmaking-from-imbalanced-observational-data(a315d279-4aac-4e0a-a5bf-649665848da1).html
dc.identifier.other PURE LINK: https://arxiv.org/abs/1904.05268
dc.identifier.other PURE LINK: https://github.com/IirisSundin/active-learning-for-decision-making
dc.identifier.other PURE LINK: http://proceedings.mlr.press/v97/sundin19a.html
dc.identifier.other PURE FILEURL: https://research.aalto.fi/files/35132269/sundin19a.pdf
dc.identifier.uri https://aaltodoc.aalto.fi/handle/123456789/39477
dc.description.abstract Machine learning can help personalized decision support by learning models to predict individual treatment effects (ITE). This work studies the reliability of prediction-based decision-making in a task of deciding which action a to take for a target unit after observing its covariates x~ and predicted outcomes p^(y~∣x~,a). An example case is personalized medicine and the decision of which treatment to give to a patient. A common problem when learning these models from observational data is imbalance, that is, difference in treated/control covariate distributions, which is known to increase the upper bound of the expected ITE estimation error. We propose to assess the decision-making reliability by estimating the ITE model’s Type S error rate, which is the probability of the model inferring the sign of the treatment effect wrong. Furthermore, we use the estimated reliability as a criterion for active learning, in order to collect new (possibly expensive) observations, instead of making a forced choice based onunreliable predictions. We demonstrate the effectiveness of this decision-making aware active learning in two decision-making tasks: in simulated data with binary outcomes and in a medical dataset with synthetic and continuous treatment outcomes. en
dc.format.mimetype application/pdf
dc.language.iso en en
dc.publisher PMLR
dc.relation.ispartof International Conference on Machine Learning en
dc.relation.ispartofseries Proceedings of the 36th International Conference on Machine Learning en
dc.relation.ispartofseries Proceedings of Machine Learning Research en
dc.relation.ispartofseries Volume 97 en
dc.rights openAccess en
dc.subject.other 113 Computer and information sciences en
dc.title Active Learning for Decision-Making from Imbalanced Observational Data en
dc.type A4 Artikkeli konferenssijulkaisussa fi
dc.description.version Peer reviewed en
dc.contributor.department Centre of Excellence in Computational Inference, COIN
dc.contributor.department Johns Hopkins University
dc.contributor.department Probabilistic Machine Learning
dc.contributor.department Department of Computer Science
dc.contributor.department Department of Computer Science en
dc.subject.keyword 113 Computer and information sciences
dc.identifier.urn URN:NBN:fi:aalto-201907304532
dc.type.version publishedVersion


Files in this item

Files Size Format View

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record

Search archive


Advanced Search

article-iconSubmit a publication

Browse

My Account