News

Happy to share that Deep Single-Index Fréchet Regression was accepted to ICML 2026!

Thrilled to share that our paper, End-to-End Deep Learning for Predicting Metric Space-Valued Outputs, has been accepted for publication in the Journal of Machine Learning Research. In this work, we introduce E2M, an end-to-end deep learning framework for predicting metric space-valued outputs via geometry-aware weighted Fréchet means.

Honored to share that our paper, Deep Fréchet Regression, has been selected as the discussion paper for the Journal of the American Statistical Association Theory & Methods Invited Session at the 2026 Joint Statistical Meetings. The invited session will take place on Monday, August 3rd (10:30am–12:20pm). As a discussion paper, our work will be accompanied by invited discussions and contributed comments, followed by a rejoinder.

I am delighted to share that two of our papers have been accepted for presentation at NeurIPS 2025.

• Fréchet Geodesic Boosting: We introduce FGBoost, a gradient boosting framework designed to intrinsically model complex regression relationships with non-Euclidean outputs in geodesic metric spaces.
• Wasserstein Transfer Learning: We propose a novel transfer learning framework for regression where outputs are probability distributions residing in the Wasserstein space.

Thrilled to announce that our paper, Deep Fréchet Regression, has been accepted for publication in the Journal of the American Statistical Association. In this work, we develop a deep learning framework for Fréchet regression, enabling flexible modeling of complex regression relationships where responses are random objects in general metric spaces.