Adversarial Vision Challenge

TL;DR

The paper discusses the NIPS 2018 Adversarial Vision Challenge, a competition aimed at advancing robust machine vision models and understanding adversarial attacks through standardized benchmarks.

Contribution

It introduces a structured competition to evaluate and improve the robustness of machine vision models against adversarial attacks.

Findings

Established benchmarks for adversarial robustness

Facilitated progress in developing resilient vision models

Identified vulnerabilities in current machine vision systems

Abstract

The NIPS 2018 Adversarial Vision Challenge is a competition to facilitate measurable progress towards robust machine vision models and more generally applicable adversarial attacks. This document is an updated version of our competition proposal that was accepted in the competition track of 32nd Conference on Neural Information Processing Systems (NIPS 2018).