Video Lectures

I will prove that the block length of every linear 3-query Locally Correctable Code (LCC) with a constant distance over any small field grows exponentially with k, the dimension of the code. This result gives the first separation between LCCs and...

I will discuss a result with Bonatti and Crovisier from 2009 showing that the C1 generic diffeomorphism f of a closed manifold has trivial centralizer; i.e. fg = gf implies that g is a power of f. I’ll discuss features of the C1 topology that enable...

Dimitroglou-Rizell-Golovko constructs a family of Legendrians in prequantization bundles by taking lifts of monotone Lagrangians. These lifted Legendrians have a Morse-Bott family of Reeb chords. We construct a version of Legendrian Contact Homology...

In this talk, I will discuss progress on showing hardness of the Minimum Circuit Size Problem (MCSP). The computational complexity of MCSP is a longstanding mystery, dating back as far as Levin's seminal work on NP-completeness in 1973. Over the...

S.T. Lee Panel Discussion: Perspectives on Democracy and Citizenship in Contemporary China
Friday, February 2, 2024
5:00 p.m. | Wolfensohn Hall 

Recent discussions about democracy in China have tackled themes relevant to its political system as well as...

I will describe the construction of a harmonic measure that reproduces a harmonic function from its Robin boundary data, which is a combination of the value of the function and its normal derivative. I shall discuss the surprising fact that this...

I will give a lightning review of results on this subject from 1988-1998 and then describe more recent developments, mostly current work in progress.  First, I will describe topological twisting of Lagrangian N=2 d=4 field theories in terms of the...

The connection between 3d pure gravity and Chern-Simons theory has been proposed and studied for decades. In this talk, I will give a brief review of the works 2304.13650 and 2401.13900 with Scott Collier and Lorenz Eberhardt, where we sharpen this...

Consider a scenario where we are learning a predictor, whose predictions will be evaluated by their expected loss. What if we do not know the precise loss at the time of learning, beyond some generic properties (like convexity)? What if the same...