Sampling Without Compromising Accuracy in Adaptive Data Analysis
[edit]
Proceedings of the 31st International Conference on Algorithmic Learning Theory, PMLR 117:297318, 2020.
Abstract
In this work, we study how to use sampling to speed up mechanisms for answering adaptive queries into datasets without reducing the accuracy of those mechanisms. This is important to do when both the datasets and the number of queries asked are very large. In particular, we describe a mechanism that provides a polynomial speedup per query over previous mechanisms, without needing to increase the total amount of data required to maintain the same generalization error as before. We prove that this speedup holds for arbitrary statistical queries. We also provide an even faster method for achieving statisticallymeaningful responses wherein the mechanism is only allowed to see a constant number of samples from the data per query. Finally, we show that our general results yield a simple, fast, and unified approach for adaptively optimizing convex and strongly convex functions over a dataset.
Related Material


