The Defense Advanced Research Projects Agency (DARPA) of the United States works primarily on the fields of research and development of Defense. It plays an important role in developing technologies that could be utilized for the military in an effective way. Now the DARPA is designing surrogate models that integrate Artificial Intelligence and Machine Learning to make it more effective.
The DARPA Ditto program develops an automated software framework that will train Machine Learning models. Engineers will find it easy to create and train Artificial Intelligence systems due to computer-generated design models. Considering the enhancements in modern Machine Learning, it is best suited for real-world functions however it comes with two limitations. They are lacking in meta-cognition and composability. Machine Learning algorithms have no knowledge of real-world functions or the represented systems. Instead, they store new information by training on additional data.
The Ditto: Intelligent Auto-Generation and Composition of Surrogate Models is responsible for creating an automated software framework. The project cost worth is estimated to be about one million dollars. The DARPA Ditto program upon creating an AI framework, makes it to learn to generate and combine surrogate models of a complex system. With the help of Microelectronics system design, it trains the Machine learning design model of subsystem components. It allows them to organize further into the levels of hierarchy. This process thus extends a helping hand to Engineers in making design-based decisions, detecting errors easily, and to reduce the risks associated with them.
The interesting thing about this program is that it simulates Integrated circuits (ICs), mixed-signal circuits boards, and network distributed systems to optimize the framework. This process makes it easier to adapt to more designs relentlessly and also assists in learning from mistakes. The project aims at generating a framework at its inception stage and then with the help of third-wave AI techniques, it helps to create a surrogate design model automatically. Once the surrogate design model is developed then it will create a proof-of-concept framework of full-system simulation. In simple terms, the project will first develop a bare-bones framework and apply AI techniques to generate Surrogate design models and enable rapid full-system simulation.
