Mathematics
Recharging Bandits
We introduce a general model of bandit problems in which the expected payout of an arm is an increasing concave function of the time since it was last played. We first develop approximation algorithms for…
Approximating General Norms by Euclidean Beyond the John’s Ellipsoid
John’s theorem proved in 1948 states that any centrally-symmetric convex body in R^d can be sandwiched by two ellipsoids up to a factor of sqrt{d}. In particular, it implies that any d-dimensional normed space embeds…
Inherent Trade-Offs in Algorithmic Fairness
Recent discussion in the public sphere about classification by algorithms has involved tension between competing notions of what it means for such a classification to be fair to different groups. We consider several of the…
On Characterizing the Capacity of Neural Networks using Algebraic Topology
The learnability of different neural architectures can be characterized directly by computable measures of data complexity. In this talk, we reframe the problem of architecture selection as understanding how data determines the most expressive and…
Dense Associative Memories and Deep Learning
Dense Associative Memories are generalizations of Hopfield nets to higher order (higher than quadratic) interactions between the spins/neurons. I will describe a relationship between these models and neural networks commonly used in deep learning. From…
I Chose STEM – Event Recap
Earlier this week Microsoft Research Montreal celebrated the International Day of Women and Girls in STEM (science, technology, engineering and mathematics) with a one-day symposium: I Chose STEM. More than 200 Canadian STEM students and…
Recent Results on Learning Filters and Style Transfer
In the first part of this talk, I will present recent results on learning image filters for low-level vision. We formulate numerous low-level vision problems (e.g., edge-preserving filtering and denoising) as recursive image filtering via…
