A University of Massachusetts, Amherst laboratory has received a contract fom the Defense Advanced Research Projects Agency to design a framework for machine learning in support of the Ditto – Intelligent Auto-Generation and Composition of Surrogate Models project.
UMass said Thursday its Biologically Inspired Neural and Dynamical Systems Lab has teamed up with Lockheed Martin to build a stucture that can automatically generate and integrate surrogate models into a single design by simulating a complex system at a rapid pace.
The partnership will use the Modular Knowledgeable AI system, a technology that features a neural compiler and a meta-cognition capability that consolidates all available knowledge.
"Meta-cognition is the ability of the human mind to leverage knowledge about the self in relation to a given task," said Hava Siegelmann, director of UMass Amherst BINDS lab.
Siegelmann added the planned framework will integrate knowledge from its own inputs and other components to perform informed computing tasks.
"We are combining our vast experience in integrated circuits design and testing with the top level of the University's machine learning neural networks to propose an automated proof-of-concept software framework for fast and accurate testing of new and updated designs," said Janet Wedgewood, lead engineer for the Ditto project at Lockheed.