Past Events

Neural networks are remarkable families of non-linear functions that form the backbone for state-of-the-art algorithms in tasks from computer vision (self-driving cars), natural language processing (Google Translate) and reinforcement learning...

I will discuss recent new developments in De Giorgi Conjecture for Allen-Cahn equation. In dimensions up to 10, we shall establish the C2,α estimates of level sets for stable solutions of Allen-Cahn. By applying reverse gluing method we show that the...

Lattice cohomology is a combinatorial invariant of isolated singularities of complex surfaces introduced by Ozsvath Stipsicz and Szabo, and independently by Nemethi. The construction, inspired by Heegaard Floer homology, was recently employed by...

This talk will focus on two topics in the foundations of neural networks: computability and false structures. In the first part I will recall briefly the concept of recursive functions and then discuss the gap between existence results, the...

A one-day workshop on Using R with Professor John Braun from the University of British Columbia (Okanagan), is the latest in the increasingly popular and high-quality series of workshops offered by the University of Calgary Biostatistics Centre (UCBC...

TBA

In this talk we will discuss a combinatorial approach to studying the geometry of the moduli space of Langlands parameters for GLn. We will first discuss several connections between the study of partitions and the study of nilpotent conjugacy classes...

A challenge in descriptive complexity is to identify logics with low complexity that simultaneously express both fundamental reachability properties and counting properties on unordered structures. In this talk, we introduce a family of logics that...

TBA

A subset of the integers P is called predictive if for all zero-entropy processes X_i; i in Z, X_0 can be determined by X_i; i in P. The classical formula for entropy shows that the set of natural numbers forms a predictive set. In joint work with...