Matthew Nelson
The Defense Advanced Research Projects Agency will host an event on Feb. 6 to familiarize interested vendors with its plan to explore methods to defend machine learning models from adversarial deception attacks.
A FedBizOpps notice posted Thursday says the proposers day will highlight DARPA's Guaranteeing AI Robustness against Deception program, which seeks to create defenses, experiments and theories that will lead to the development of deception-resistant algorithms and systems.
The agency noted it looks to apply prediction, location and detection functions in efforts to shield machine learning models from a range of digital modalities.
The event will feature a summary presentation as well as a question -and-answer session to accommodate queries regarding the project.
Potential attendees should register via email no later than Feb. 1.
