BARC talk by Kasper Green Larsen, Sep 2020

Thursday, 3 September 2020, Kasper Green Larsen, Associate Professor at the University of Aarhus, will give a talk on "Near-Tight Margin-Based Generalization Bounds for Support Vector Machines".

Title:
Near-Tight Margin-Based Generalization Bounds for Support Vector Machines

Abstract:

Support Vector Machines (SVMs) are among the most fundamental tools for binary classification. In its simplest formulation, an SVM produces a hyperplane separating two classes of data using the largest possible margin to the data. The focus on maximizing the margin has been well motivated through numerous generalization bounds. In this talk, we revisit and improve the classic generalization bounds in terms of margins. Furthermore, we complement our new generalization bound by a nearly matching lower bound, thus almost settling the generalization performance of SVMs in terms of margins.

This is joint work with Allan Grønlund and Lior Kamma from Aarhus University, and it was published at ICML'20.

Bio:
Kasper Green Larsen is an Associate Professor at the Department of Computer Science at Aarhus University. His research spans many areas of theoretical computer science, but in particular he enjoys working on data structures, range searching, lower bounds, dimensionality reduction, and streaming algorithms.

Kasper received his Ph. D. from Aarhus University in 2013 and since 2017 he has been a member of the Young Academy under the Royal Danish Academy of Sciences and Letters.

Read more about Kasper Green Larsen