Harnessing AI for Development and Evaluation of Systems (HADES)

MTSI supports the Federal Gov’t by leveraging our MTSI-developed HADES Tool that verifies real world performance and improves the robustness and reliability of Machine Learning/AI-based algorithms. The HADES ‘Adversary’ algorithm learns the performance space of a potential classification algorithm that could be fielded. The Adversary learns performance bounds by changing the physical dimensions of the threat objects the classifier trained against and calling the MTSI High Rate Thermal and Signature Solver (HiRTSS) to create new data to present to the classifier. This creates new data sets which can be used to improve the classifier and make it more robust in real-world situations.