3 papers
cs.LG2026
Belted Engression: Sufficient Dimension Reduction for Generative Distributional Regression
Wenxi Tan, Bing Li, Lingzhou Xue
Modern conditional generative models face significant challenges when learning complex covariate dependencies. While sufficient dimension reduction (SDR) provides a principled appr…
stat.ME2026
Distance Profile Embedding for Independence and Conditional Independence Testing of Random Objects
Wenxi Tan, Bing Li, Lingzhou Xue
Testing independence or conditional independence is fundamental to statistical inference, yet existing methods for non-Euclidean random objects often face a difficult trade-off bet…
stat.ME2026
A Unified Framework for Nonlinear Mediation Analysis of Random Objects
Wenxi Tan, Bing Li, Lingzhou Xue
Mediation analysis for complex, non-Euclidean data, such as probability distributions, compositions, images, and networks, presents significant methodological challenges due to the…