Event Type: Talk

Abstract:

First-order methods are widely used in optimization and machine learning, and their behavior is often
analyzed through the spectrum of worst case convergence rates. Obtaining such guarantees is often
difficult and both time consuming and error-prone. Starting with the work of Drori and Teboulle (2014),
novel techniques have been used to gain numerical insights, leading to the release of various
performance estimation (PE) software.
In this talk, I will show how various computer-aided techniques can be used to study first-order
optimization methods in a systematic way. From performance estimation problems with automated
Lyapunov discovery, to symbolic regression and computer algebra systems, novel tools completely
reshape the way we approach theory of optimization.
As a main example, I will focus on error feedback methods used with compressed communication in
distributed optimization. While error feedback has been widely studied, existing theory often provides
untight (thus unreliable) bounds. I will present tight analyses with matching lower bounds that allow a fair
comparison between error feedback schemes and standard compressed gradient descent, and help
explain when error feedback is useful and when it is not.
Overall, the talk aims to show how various computer-aided proofs can lead to clearer and more reliable
insights into first-order optimization methods.

Abstract:

Advances in spatial and multiplexed imaging technologies are revealing the intricate organization of human tissues at single-cell and molecular resolution. However, transforming these rich datasets into biological and clinical insight requires new computational paradigms. In this talk, I will present how modern AI and machine learning methods, ranging from classical machine learning models to generative AI, are redefining the analysis of 2D and 3D spatial omics data. More importantly, I will illustrate how these approaches enable functional cell type discovery, cancer subgroup identification, and in silico spatial experimental design, translating data into actions for precision medicine.

Dr. Yuqi Tan, Instructor (Microbiology & Immunology, a non-tenured faculty track) at Stanford University, holds a BSc in Cell and Molecular Biology from the Chinese University of Hong Kong and a PhD in Computational Biology from Johns Hopkins University. With a record of more than 20 scholarships and awards, she specializes in crafting computational tools that integrate and transfer knowledge across various single-cell data modalities, whether in two or three dimensions. Her expertise finds practical applications in determining cell type identities and elucidating the spatial orchestration of cell types, especially within the domains of stem cell engineering, cancer immunotherapy, cancer initiation, and psychiatric diseases.

Website: https://yuqiyuqitan.github.io/