To analyze trends in things like wind speed and ocean temperatures, scientists often use datasets where each observation is a function—like a curve or time series—instead of a single value. Current tools struggle with this sort of complex data, so SCI Ph.D. student Wenzheng Tao, faculty member Sarang Joshi, and emeritus faculty member Ross Whitaker developed a better method: Bayesian Scalable Functional Data Analysis. It automatically simplifies high-dimensional data while retaining key details, making it useful across many scientific fields. “Imagine recording the temperature at every depth in the ocean, every day, across the entire globe. That’s billions of numbers,” Tao said. “Our method compresses that information to a handful of ‘summary waves’ you can store and explore on a regular laptop. The same trick works for personal health data, wind-farm planning, even finance—anywhere you have messy curves instead of simple spreadsheets.”
Algorithms has a 37% acceptance rate and publishes monthly. Each issue contains about 60 articles, only one of which is selected for the cover. Tao was excited for the recognition. “Being chosen is a nice bit of external validation and visibility,” he said. “A cover helps the work reach a broader audience, which is especially beneficial considering its cross-disciplinary applications. I’m grateful to my advisors, Prof. Sarang Joshi and Prof. Ross Whitaker, for the guidance that made this possible, and to the Algorithms editors for the nod.”
Read the SCI team's full paper.