The goal is to perform the minimization based only on samplesS = {z1,...,zn}drawn i.i.d.

This paper considers the problem of finding the (δ,ǫ)-Goldstein stationary point of the Lipschitz continuous objective, which is a rich funct ion class to cover a large number of important applications. We construct a nove l zeroth-order quantum estimator for the gradient of the smoothed surrogate.

In this paper, we use masked autoencoders for this one-sample learning problem.

Neural Information Processing Systems http://nips.cc/