Accession Number:

AD1088314

Title:

Measuring and Comparing Robustness of ML Algorithms Under Adversarial Attack

Descriptive Note:

Technical Report

Corporate Author:

CARNEGIE-MELLON UNIV PITTSBURGH PA PITTSBURGH United States

Personal Author(s):

Report Date:

2017-08-01

Pagination or Media Count:

6.0

Abstract:

A machine learning algorithm can be evaluated for robustness against any number of different types of attacks. We consider attacks that seek to manipulate the training andor testing data inputs to a machine learning algorithm. Specifically, we do not consider physical attacks on machines hosting the algorithm.

Subject Categories:

  • Computer Programming and Software

Distribution Statement:

APPROVED FOR PUBLIC RELEASE