Detailed cosmological observations suggest invisible substances dominate the energy budget of the Universe and play an outsized role in its growth from the Big Bang to today. Theories put forward to explain the fundamental nature of these “dark” components require time-consuming numerical computations to test their predictions against observations from state-of-the-art telescopes. Recent advances in artificial intelligence and machine learning (AI/ML) dramatically speed up such computation, making it possible to explore efficiently a broader range of plausible theories. This project develops AI/ML-based computational tools enabling novel searches for the physical origin and properties of the invisible elements making up the cosmos. These tools leverage complementarity between astronomical observations and laboratory-based experiments in determining the physics governing the growth of the Universe. In parallel, a pilot program is established to mentor and engage students early-on in research. This program focuses on acquisition of transferable skills broadly applicable to careers in STEM, while providing students with a support community informing their sense of relevance within physics. The project aims to determine whether neutrinos have nonstandard interactions that drive cosmic expansion and the growth of large-scale structure. To do so, this project incorporates the latest advances in AI/ML to develop a fully differentiable cosmological pipeline that dramatically speed