Research Interests
Applied and Computational Mathematics / Data Science
Research Statement
I develop mathematically provable machine learning algorithms to circumvent prohibitively costly computations in scientific computing, thereby opening new avenues for scientific breakthroughs. Modern scientific advances in peta and exascale computing and high throughput technologies are enabling for the first time true scale bridging research in numerous computational fields such as quantum chemistry, materials science, computational biology, astrophysics, climate science, and fluid mechanics. These advances have lead to massive amounts of high dimensional distributed data, and interpreting and analyzing this data is a fundamental problem facing science. Machine learning is one approach to extract information from these data sets. However, to be able to apply machine learning algorithms to interpret these multiscale data sets, new paradigms are needed that are capable of learning intrinsic variables that encode complex interactions across scales.
