Novel AI approaches inspired by fundamental physics are reshaping computational methods in nuclear theory. In this talk, I’ll discuss opportunities for machine learning—particularly generative models—to accelerate first-principles lattice quantum field theory calculations in particle and nuclear physics. Challenges include embedding complex (gauge) symmetries into model architectures and scaling models to the many degrees of freedom in state-of-the-art numerical studies. Proof-of-principle results will show that generative model sampling can be orders of magnitude more efficient than traditional Hamiltonian/hybrid Monte Carlo approaches, with a discussion of future prospects.
This event will take place in PAB C-520. All interested graduate students and faculty are invited to attend.
Participants are also welcome to join via Zoom. Zoom link will be available via announcement email, or by contacting intmail[at]uw.edu