Gaussian Mixture Counterfactual Generator
Keywords: Gaussian mixture model, synthetic data generation, clinical trial, individual treatment effect, counterfactual generation
Abstract: Generating synthetic control arms is a key challenge in clinical research, particularly in crossover trials where placebo data becomes unavailable after patients switch to active treatment. The absence of placebo data complicates estimating long-term efficacy and safety. To solve this, we propose a Gaussian mixture model that generates counterfactual data without needing control data for training. This method handles time-varying, continuous doses and estimates effects between treatment switchers and an extended placebo group, providing valuable insights for treatment effects, evidence generation, and decision-making.
Primary Area: causal reasoning
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 12561
Loading