Machine learning-based system to improve data sharing for DARPA programNews
February 02, 2021
ARLINGTON, Va. BAE Systems’ FAST Labs research and development organization has been tapped by the Defense Advanced Research Projects Agency (DARPA) to develop a scalable machine learning system designed to provide data anonymity to improve data sharing. The program, called Cooperative Secure Learning (CSL), has potential cybersecurity applications.
DARPA asked BAE Systems to develop a scalable machine learning solution that preserves the privacy of the data and the model, intended to enable Cooperative Secure Learning.
The company's Privacy-preserving Arithmetic Computation for Encrypted Learning solution, known as PArCEL solution, is designed to mitigate common privacy challenges by combining recent research in cooperative learning on encrypted feature embeddings with new network log sanitization techniques.
Unlike other approaches such as encryption of raw private data, BAE Systems claims that the company's focus on feature embeddings could reduce computational complexity while providing additional protection against information leakage and reduced information sharing.
Work on this approximately $1 million program, which includes research teammates led by Prof. Khorrami at New York University, adds to BAE Systems cybersecurity and machine learning portfolios.