Program Summary
The Physics of Artificial Intelligence (PAI) program is part of a broad DAPRA initiative to develop and apply “Third Wave” AI technologies to sparse data and adversarial spoofing, and that incorporate domain-relevant knowledge through generative contextual and explanatory models.
It is anticipated that AI will play an ever larger role in future Department of Defense (DoD) activities, ranging from scientific discovery, to real-time sensor processing, to control and coordination of composable systems. However, despite rapid progress of AI in the commercial sector – particularly in the subfield of machine learning – AI’s successful integration into numerous DoD applications has proven challenging. Key challenges include the development of causal, predictive models and dealing with incomplete, sparse, and noisy data.
To facilitate better incorporation of AI into DoD systems, the PAI program is exploring novel AI architectures, algorithms, and approaches that “bake in” physics, mathematics, and prior knowledge relevant to DoD application domains. PAI aims to show that embedding physics and prior knowledge into AI will help to overcome the challenges of sparse data and will facilitate the development of generative models that are causal and explanative.