We study the function space characterization of the inductive bias resul...
We investigate the problem of efficiently computing optimal transport (O...
Adaptive sampling is a useful algorithmic tool for data summarization
pr...
Random binning features, introduced in the seminal paper of Rahimi and R...
Randomized smoothing is a recently proposed defense against adversarial
...
The Optimal Transport (a.k.a. Wasserstein) distance is an increasingly
p...
Recent works have shown the effectiveness of randomized smoothing as a
s...
We present new secure protocols for approximate k-nearest neighbor searc...
We study the problem of mean estimation for high-dimensional distributio...
Most of the efficient sublinear-time indexing algorithms for the
high-di...
Space partitions of R^d underlie a vast and important class of
fast near...
In our recent work (Bubeck, Price, Razenshteyn, arXiv:1805.10204) we arg...
We introduce and study the notion of an outer bi-Lipschitz extension of ...
Consider an instance of Euclidean k-means or k-medians clustering. We
sh...
The nearest neighbor problem is defined as follows: Given a set P of n
p...
Why are classifiers in high dimension vulnerable to "adversarial"
pertur...
We introduce a new distance-preserving compact representation of
multi-d...