Country
The Discrete Gaussian for Differential Privacy
Work done while at IBMResearch - Almaden.
Unified Generative and Discriminative Training for Multi-modal Large Language Models Wei Chow
In recent times, Vision-Language Models (VLMs) have been trained under two predominant paradigms.
Efficient Aggregated Kernel Tests using IncompleteU-statistics
Theminimax (i.e.optimal) rateoverthe SobolevballSsd(R) is N 2s/(4s+d) forthetwo-sample (Liand Yuan, 2019, Theorem 5 (ii)) andindependence (Albertetal., 2022, Theorem 4; Berrett etal., 2021, Corollary 5) problems.
b4edda67f0f57e218a8e766927e3e5c5-Paper.pdf
The priors used in this presentation include variants oftotalvariation, Laplacian regularization, bilateral filtering, sparse coding on learned dictionaries, and non-local self similarities. Our models are fully interpretable as well as parameter and data efficient.
Statistical Limits of Adaptive Linear Models: Low-Dimensional Estimation and Inference
Estimation and inference in statistics pose significant challenges when data are collected adaptively.