Professor Prasanth Nair of the University of Toronto Institute for Aerospace Studies in the Faculty of Applied Science & Engineering has received a Connaught Innovation Award to support his computational framework, known as Deep Bayesian Analytics Engine (DBAE).
“In collaboration with my graduate student, Trefor Evans, I developed a novel, deep learning architecture coupled to a Bayesian analytics engine, which together forms a powerful and highly flexible computational framework for decision support in complex engineering applications,” says Nair, a Canada Research Chair in Computational Modeling and Design Optimization Under Uncertainty.
“Our particular focus is on high-dimensional problems that are common in the aerospace industry; however, the core platform could be adapted to any sector that requires integrating diverse sources of data to make optimal decisions.”
The DBAE has enabled Nair’s team to dramatically speed up tasks that routinely arise in the computation of decision analytics, he says, allowing their prototype software to provide “rigorous decision-making support” that wouldn’t be possible using standard techniques.
“The Connaught Innovation Award will enhance our prototype software implementation and allow us to carry out additional testing, and commercialize the technology with potential end-users,” Nair says. “To enhance our prototype software and make it attractive to users in industry, our main objectives are to improve the user interface and visualization capabilities; validate the software on a comprehensive set of challenging benchmarks; increase the optimization capabilities; prepare the software for integration with third-party simulation tools; and solicit and incorporate feedback from user testing.”