NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:5042
Title:Tight Dimensionality Reduction for Sketching Low Degree Polynomial Kernels

This paper presents tight bounds on the dimension of random projection for tensor product of vectors, achieving exponential improvement on sketch dimension compared to the prior work. Their method also enjoyes efficient computation. The reviewers found the work solid and of high significance.