|Machine Learning for Quantum Systems|
|Samuel Kellar, Louisiana State University|
|Digital Media Center 1008B
May 28, 2019 - 03:30 pm
The Soft Gap Anderson model, with a hybridization function proportional to omega^r, serves as a simple test for machine learning. A combination of supervised and unsupervised methods learn directly on the Hirsch Fye Quantum Monte Carlo decoupled fields and separate the into two phases near the predicted value of r=0.5
Samuel Kellar Graduated with a Bachelor of Science in Physics from Brigham Young University. As a graduate student he uses a dynamical cluster approximation to study the Hubbard model in 3 dimensions. He worked with the Ste||ar at the Center for Computation & Technology in improving efficiency of highly parallel quantum calculations.