Lone Star Section Hosted Secure Development of Machine Learning Against Poisoning Attacks,
SwRI’s Dr. Garrett Jares presented "Secure Development of Machine Learning Against Poisoning Attacks,” at the Southwest Research Institute on Thursday, March 14, at noon. The event is hosted by the IEEE’s Lone Star Section and co-sponsored by IEEE’s Women in Engineering.
Abstract
Recent research has revealed that machine learning models are vulnerable to adversarial attacks that seek to manipulate the model to induce undesired behavior or extract sensitive information. Such vulnerabilities are particularly concerning for the use of these methods in aerospace and defense applications where safety and security are paramount. This research evaluates an approach that combines several attack detection methods in tandem to produce an intrusion detection system (IDS) that ensures security at each stage of the model’s lifecycle. The performance of the pipelined detection approach is compared to the performance of each individual detector with the hypothesis that the Combined IDS will result in improved security. The goal of this research is to move toward a practice for secure AI development and operation.