SurvITE: Learning Heterogeneous Treatment Effects from Time-to-Event Data
- Authors
- Curth, Alicia; Lee, Changhee; van der Schaar, M.
- Issue Date
- Dec-2021
- Publisher
- Neural information processing systems foundation
- Citation
- Advances in Neural Information Processing Systems, v.32, pp 26740 - 26753
- Pages
- 14
- Journal Title
- Advances in Neural Information Processing Systems
- Volume
- 32
- Start Page
- 26740
- End Page
- 26753
- URI
- https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/61934
- DOI
- 10.48550/arXiv.2110.14001
- ISSN
- 1049-5258
- Abstract
- We study the problem of inferring heterogeneous treatment effects from time-to-event data. While both the related problems of (i) estimating treatment effects for binary or continuous outcomes and (ii) predicting survival outcomes have been well studied in the recent machine learning literature, their combination - albeit of high practical relevance - has received considerably less attention. With the ultimate goal of reliably estimating the effects of treatments on instantaneous risk and survival probabilities, we focus on the problem of learning (discrete-time) treatment-specific conditional hazard functions. We find that unique challenges arise in this context due to a variety of covariate shift issues that go beyond a mere combination of well-studied confounding and censoring biases. We theoretically analyse their effects by adapting recent generalization bounds from domain adaptation and treatment effect estimation to our setting and discuss implications for model design. We use the resulting insights to propose a novel deep learning method for treatment-specific hazard estimation based on balancing representations. We investigate performance across a range of experimental settings and empirically confirm that our method outperforms baselines by addressing covariate shifts from various sources.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - College of Software > Department of Artificial Intelligence > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/61934)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.