Deep Knockoffs
This provides a flexible, model-free tool for controlled variable selection in statistics, addressing a bottleneck in high-dimensional data analysis, though it builds incrementally on the existing model-X framework.
The paper tackles the problem of generating model-X knockoffs for arbitrary data distributions without specifying the underlying model, using deep generative models to iteratively optimize a validity criterion based on pairwise exchangeability. The method is validated through extensive numerical experiments and applied to identify mutations linked to drug resistance in HIV, demonstrating its generality and effectiveness.
This paper introduces a machine for sampling approximate model-X knockoffs for arbitrary and unspecified data distributions using deep generative models. The main idea is to iteratively refine a knockoff sampling mechanism until a criterion measuring the validity of the produced knockoffs is optimized; this criterion is inspired by the popular maximum mean discrepancy in machine learning and can be thought of as measuring the distance to pairwise exchangeability between original and knockoff features. By building upon the existing model-X framework, we thus obtain a flexible and model-free statistical tool to perform controlled variable selection. Extensive numerical experiments and quantitative tests confirm the generality, effectiveness, and power of our deep knockoff machines. Finally, we apply this new method to a real study of mutations linked to changes in drug resistance in the human immunodeficiency virus.