Selected Talks

Benefits of Convolutional Models: A Kernel Perspective. UMN 2022, Harvard 2022, Gatsby 2022, BIRS 2022, Flatiron PhysML 2022.

On the Sample Complexity of Learning under Invariance and Geometric Stability. Berkeley 2021, Flatiron 2021, Johns Hopkins 2021, Cambridge 2022.

Approximation and Generalization with Deep Convolutional Kernels. EPFL, 2021.

Foundations of Deep Convolutional Models through Kernel Methods. Phd thesis 2019, MSR NY 2020, TTIC 2020, NYU CDS 2020 (video), MALGA 2021 (video)

On the Inductive Bias of Neural Tangent Kernels. GIPSA-lab, Grenoble, 2019.

Invariance and Stability to Deformations of Deep Convolutional Representations. NYU CILVR 2017, SMAI-MODE 2018, ENS 2018, TTIC 2019, MSR 2019 (video), UC Berkeley 2019

A Contextual Bandit Bake-Off. Telecom ParisTech, 2018.

A Family of Stochastic Surrogate Optimization Algorithms. SIAM Optimization, Vancouver, 2017.

Stochastic Optimization with Variance Reduction for Infinite Datasets with Finite Sum Structure. MACARON workshop, Grenoble, 2017.

An Online EM Algorithm in Hidden (semi-)Markov Models for Audio Segmentation and Clustering. ICASSP, Brisbane, 2015.

Online Learning for Audio Clustering and Segmentation. IRCAM, Paris, 2014.