Washington State University April 3 announced that an Indian American researcher at its university has been honored with an Early Career award from the National Science Foundation.
Berry Family Professor in the School of Electrical Engineering and Computer Science Janardham Rao Doppa was the recipient of a CAREER award which will grant him $550,000 over five years.
The money will provide significant research support as Doppa works to develop algorithms to support scientists and engineers in their decision-making and experimental design process, the university said.
In many science and engineering fields, from computer chip design to discovering materials and studying microbiomes, problem solving is complex and multi-faceted, it said.
Understanding complex systems with many variables combined with experimental data often include too many factors for humans to consider. Scientists often have to rely on intuition to make their decisions.
“A lot of engineers and scientists rely on intuition and what worked or didn’t work as they develop an experiment,” said Doppa in the university report. “There isn’t a formal model to guide that process.”
So researchers combine their domain knowledge and data from past experiments to develop a sequence of experiments to achieve their objectives using minimal resources.
“In all of these problems, the design space is very large, and each design is a discrete combinatorial structure or a hybrid structure with a mixture of discrete and continuous design variables,” he said.
The CAREER award will support Doppa’s work to develop general-purpose learning and reasoning algorithms to support engineers and scientists, and to streamline experimental processes, the university said.
He will collaborate with scientists to apply the algorithms, in particular, in the fields of computer hardware, materials, and synthetic microbiomes, helping the researchers decide on the optimal way to conduct their experiments, the report said.
The work builds on his success in developing algorithms for electronic design automation and computer architecture domains.
Doppa received the Voiland College’s Reid Miller teaching excellence award in 2018, and has worked to democratize computer science and machine learning skills for interdisciplinary research.
He will be developing a short course on data-driven design optimization and will be working with researchers to develop efficient algorithms with an easy interface. While he is teaching them, he will also be learning and working to improve the easy-to-use algorithms, it said.
“Scientists and engineers will be specifying the ‘what’ part of the design optimization problem and the algorithms will take care of the `how’ part,” said Doppa. “The tools need to be easy to use.”