Technical Notes and Slides


Sampling Methods for the von Mises-Fisher Distributions and Smoothed Bootstrap

Shall We Always Avoid Overfitting? – A generalized framework of the classical bias-variance trade-off in modern deep learning regime (Slides for a brief introduction to the double-descent phenomenon)

Hitting Distribution of 2-dimensional Brownian Motions on a Wedge

Conditional Quantile Regression With Applications to User-Preferred Price Prediction (Slides for applying conditional quantile regression and conformal inference in a recommendation setting)

Paper Reading Notes and Other Presentation Slides

1. Robust Optimization and Inference on Manifolds (by Lizhen Lin, Drew Lazar, Bayan Sarpabayeva, David B. Dunson):

2. An Efficient and Continuous Voronoi Density Estimator (by Giovanni Luca Marchetti, Vladislav Polianskii, Anastasiia Varava, Florian T. Pokorny, Danica Kragic):

3. Smoothed Nonparametric Derivative Estimation using Weighted Difference Quotients (by Yu Liu and Kris De Brabanter):

4. A Family of Density-Scaled Filtered Complexes (by Abigail Hickok):

5. Synthetic Multimodal Data Modelling for Data Imputation (by Francisco Carrillo-Perez, Marija Pizurica, Kathleen Marchal, and Olivier Gevaert):