Speaker: Dr. Eran Kaufman
Title: Nested Barycentric
Coordinate System as an Explicit Feature Map
Zoom Link
Abstract:
We introduce a new embedding technique based on a barycentric coordinate system. We show that our embedding can be used to transform the problem of polytope approximation into one of finding a linear classifier in a higher dimensional (but nevertheless quite sparse) representation. In effect, this embedding maps a piecewise linear function into an everywhere-linear function and allows us to invoke well-known algorithms for the latter problem to solve the former.
We demonstrate that our embedding has applications to the problems of approximating separating polytopes — in fact, it can approximate any convex body and unions of convex bodies — as well as to classification by separating polytopes and piecewise linear regression.
Bio:
Eran Kaufman is currently a lecturer at Ariel University, Rupin, Shenkar, and Afeka academy.
He received his B.Sc. from the Technion Institute and M.S.c from the Tel Aviv University in electrical engineering.
He worked at Nokia as a senior programmer, Ceragon networks as the Embedded team Leader, and ARM as a System and Software Architect.
He then continued to receive his Ph.D. degree from Ariel University in Mathematics and Computer Sciences, and he is currently a Postdoc at Ben-Guruin under the supervision of Arye
Kontorowich.
He also works with Google as a Machine learning advisor for the industry.
His research interests lie in Machine learning, Cybersecurity, and computer vision.
