Abstract

Data science in low-dimensional spaces is motivated by applications in mapping, navigation, geographic resource allocation, modeling of body shapes and chemical structures, and more. In addition to datasets that naturally reside in low-dimension spaces, dimension-reduction methods can often transform high dimensional data to lower-dimensional data while preserving properties of interest. Since many computational problems are intractable for high-dimensional data but potentially tractable for low-dimensional data, it is useful to establish the algorithmic foundations of data science on low-dimensional data, to understand the special properties of such data, and to identify computational methods that are highly effective when applied to such data.

This workshop will bring together researchers in academia and industry to explore algorithmic and data analysis technique specialized for low-dimensional data, and application areas in which such problems arise. The focus of this workshop is on algorithms with provable performance guarantees. In particular, we are interested in algorithms that scale to large datasets, and algorithms that adapts to distributed environments and to online settings.

This workshop is part of the Brown Data Science Initiative project, funded by the National Science Foundation's TRIPODS program.