Search Results for author: Matt Jacobs

Found 4 papers, 3 papers with code

An Optimal Transport Approach for Computing Adversarial Training Lower Bounds in Multiclass Classification

1 code implementation17 Jan 2024 Nicolas Garcia Trillos, Matt Jacobs, Jakwang Kim, Matthew Werenski

Recent works have developed a connection between AT in the multiclass classification setting and multimarginal optimal transport (MOT), unlocking a new set of tools to study this problem.

It begins with a boundary: A geometric view on probabilistically robust learning

1 code implementation30 May 2023 Leon Bungert, Nicolás García Trillos, Matt Jacobs, Daniel Mckenzie, Đorđe Nikolić, Qingsong Wang

Although deep neural networks have achieved super-human performance on many classification tasks, they often exhibit a worrying lack of robustness towards adversarially generated examples.

On the existence of solutions to adversarial training in multiclass classification

no code implementations28 Apr 2023 Nicolas Garcia Trillos, Matt Jacobs, Jakwang Kim

We study three models of the problem of adversarial training in multiclass classification designed to construct robust classifiers against adversarial perturbations of data in the agnostic-classifier setting.

Binary Classification

The Multimarginal Optimal Transport Formulation of Adversarial Multiclass Classification

1 code implementation27 Apr 2022 Nicolas Garcia Trillos, Matt Jacobs, Jakwang Kim

We study a family of adversarial multiclass classification problems and provide equivalent reformulations in terms of: 1) a family of generalized barycenter problems introduced in the paper and 2) a family of multimarginal optimal transport problems where the number of marginals is equal to the number of classes in the original classification problem.

Binary Classification Classification +1

Cannot find the paper you are looking for? You can Submit a new open access paper.