The Everlasting Database: Statistical Validity at a Fair Price

03/12/2018
by   Blake Woodworth, et al.
0

The problem of handling adaptivity in data analysis, intentional or not, permeates a variety of fields, including test-set overfitting in ML challenges and the accumulation of invalid scientific discoveries. We propose a mechanism for answering an arbitrarily long sequence of potentially adaptive statistical queries, by charging a price for each query and using the proceeds to collect additional samples. Crucially, we guarantee statistical validity without any assumptions on how the queries are generated. We also ensure with high probability that the cost for M non-adaptive queries is O( M), while the cost to a potentially adaptive user who makes M queries that do not depend on any others is O(√(M)).

READ FULL TEXT

page 1

page 2

page 3

page 4

research
02/17/2023

Subsampling Suffices for Adaptive Data Analysis

Ensuring that analyses performed on a dataset are representative of the ...
research
12/19/2017

Calibrating Noise to Variance in Adaptive Data Analysis

Datasets are often used multiple times and each successive analysis may ...
research
05/24/2023

Adaptive Data Analysis in a Balanced Adversarial Model

In adaptive data analysis, a mechanism gets n i.i.d. samples from an unk...
research
04/01/2020

Statistical Queries and Statistical Algorithms: Foundations and Applications

We give a survey of the foundations of statistical queries and their man...
research
02/13/2016

A Minimax Theory for Adaptive Data Analysis

In adaptive data analysis, the user makes a sequence of queries on the d...
research
09/14/2018

The Generic Holdout: Preventing False-Discoveries in Adaptive Data Science

Adaptive data analysis has posed a challenge to science due to its abili...
research
12/17/2019

Mosaic: A Sample-Based Database System for Open World Query Processing

Data scientists have relied on samples to analyze populations of interes...

Please sign up or login with your details

Forgot password? Click here to reset