Post-doctoral research associate Devanshu Agrawal received a 2022 National Science Foundation (NSF) Mathematical Sciences Postdoctoral Research Fellowship (MSPRF). He plans to apply the fellowship to his work here at UT with Associate Professor Jim Ostrowski in the Department of Industrial and Systems Engineering.
Investigations with Ostrowski provided a springboard toward the proposed research for the award.
“We developed a deep neural network architecture called a ‘group-equivariant autoencoder’ that could successfully identify spontaneous symmetry breaking in certain toy models of physical systems,” said Agrawal.
This fellowship will enable further inquiry into this phenomena.
“Our proposed research will be driven by the question: How can deep neural networks be used to identify spontaneous symmetry breaking in simulated quantum materials?” said Agrawal. “We are particularly interested in the mathematics underpinning this question, such as combinatorial optimization over symmetry-constrained neural networks.”
Writer: Randall Brown