Gradient Compressed Sensing: A Query-Efficient Gradient Estimator for High-Dimensional Zeroth-Order Optimization

We study nonconvex zeroth-order optimization (ZOO) in a high-dimensional space R d for functions with approximately s-sparse gradients. To reduce the dependence on the dimensionality d in the query complexity, high-dimensional ZOO methods seek to leverage gradient sparsity to design gradient estimators. The previous best method needs O s log d s queries per step to achieve O 1 T rate of convergence w.r.t. the number T of steps. In this paper, we propose Gradient Compressed Sensing (GraCe), a query-efficient and accurate estimator for sparse gradients that uses only O s log log d s queries per step and still achieves O 1 T rate of convergence. To our best knowledge, we are the first to achieve a doublelogarithmic dependence on d in the query complexity, and our proof uses weaker assumptions than previous work. Our proposed GraCe generalizes the Indyk-Price-Woodruff (IPW) algorithm in compressed sensing from linear measurements to nonlinear functions. Furthermore, since the IPW algorithm is purely theoretical due to its impractically large constant, we improve the IPW algorithm via our dependent random partition technique together with our corresponding novel analysis and successfully reduce the constant by a factor of nearly 4300. Our GraCe is not only theoretically query-efficient but also achieves strong empirical performance. We benchmark our GraCe against 12 existing ZOO methods with 10000-dimensional functions and demonstrate that GraCe significantly outperforms existing methods. Our code is publicly available at https://github.com/q-rz/ ICML24-GraCe .