Fall 2021

Algorithmic Advances for Statistical Inference with Combinatorial Structure

Oct 11, 2021 to Oct 15, 2021 

Add to Calendar


Laurent Massoulie (Microsoft Research-INRIA Joint Centre; chair), Po-Ling Loh (University of Wisconsin, Madison), Jiaming Xu (Duke University)

The theme of this workshop is the interplay between problem structure and computational complexity, combining the strength of the statistical and algorithmic mindsets. The focus will be on understanding how algorithms can exploit problem structure, and on understanding which tools in our algorithmic toolkit are suited for different structured inference tasks. 

The program will feature surprising and deep new algorithmic insights for prominent specific problems, such as graph matching, learning Gaussian graphical models, optimization in spin glasses, and more. At the same time, the program will highlight the broader emerging understanding of the power of classes of algorithms (such as gradient descent, message passing, generalized belief propagation, and convex programs) for families of structured problems. 

All events take place in the Calvin Lab auditorium.

Further details about this workshop will be posted in due course. Enquiries may be sent to the organizers workshop-si2 [at] (at this address).