English
Related papers

Related papers: Generalization for Adaptively-chosen Estimators vi…

200 papers

Adaptivity is an important feature of data analysis---the choice of questions to ask about a dataset often depends on previous interactions with the same dataset. However, statistical validity is typically studied in a nonadaptive model,…

Machine Learning · Computer Science 2015-11-10 Raef Bassily , Kobbi Nissim , Adam Smith , Thomas Steinke , Uri Stemmer , Jonathan Ullman

Datasets are often used multiple times and each successive analysis may depend on the outcome of previous analyses. Standard techniques for ensuring generalization and statistical validity do not account for this adaptive dependence. A…

Machine Learning · Computer Science 2018-06-13 Vitaly Feldman , Thomas Steinke

Overfitting is the bane of data analysts, even when data are plentiful. Formal approaches to understanding this problem focus on statistical inference and generalization of individual analysis procedures. Yet the practice of data analysis…

Machine Learning · Computer Science 2015-09-28 Cynthia Dwork , Vitaly Feldman , Moritz Hardt , Toniann Pitassi , Omer Reingold , Aaron Roth

Adaptivity is an important feature of data analysis---typically the choice of questions asked about a dataset depends on previous interactions with the same dataset. However, generalization error is typically bounded in a non-adaptive…

Machine Learning · Computer Science 2015-11-11 Raef Bassily , Adam Smith , Thomas Steinke , Jonathan Ullman

We present a new adaptive algorithm for learning discrete distributions under distribution drift. In this setting, we observe a sequence of independent samples from a discrete distribution that is changing over time, and the goal is to…

Machine Learning · Computer Science 2024-03-11 Alessio Mazzetto

Adaptive data analysis has posed a challenge to science due to its ability to generate false hypotheses on moderately large data sets. In general, with non-adaptive data analyses (where queries to the data are generated without being…

Methodology · Statistics 2018-09-18 Preetum Nakkiran , Jarosław Błasiok

A great deal of effort has been devoted to reducing the risk of spurious scientific discoveries, from the use of sophisticated validation techniques, to deep statistical methods for controlling the false discovery rate in multiple…

Machine Learning · Computer Science 2016-03-03 Cynthia Dwork , Vitaly Feldman , Moritz Hardt , Toniann Pitassi , Omer Reingold , Aaron Roth

We study the fundamental task of estimating the median of an underlying distribution from a finite number of samples, under pure differential privacy constraints. We focus on distributions satisfying the minimal assumption that they have a…

Statistics Theory · Mathematics 2020-11-13 Christos Tzamos , Emmanouil-Vasileios Vlatakis-Gkaragkounis , Ilias Zadik

In statistical learning, a dataset is often partitioned into two parts: the training set and the holdout (i.e., testing) set. For instance, the training set is used to learn a predictor, and then the holdout set is used for estimating the…

Machine Learning · Computer Science 2019-11-05 Jun Zhao

We give a new proof of the "transfer theorem" underlying adaptive data analysis: that any mechanism for answering adaptively chosen statistical queries that is differentially private and sample-accurate is also accurate out-of-sample. Our…

Machine Learning · Computer Science 2024-06-05 Christopher Jung , Katrina Ligett , Seth Neel , Aaron Roth , Saeed Sharifi-Malvajerdi , Moshe Shenfeld

Sample average approximation (SAA), a popular method for tractably solving stochastic optimization problems, enjoys strong asymptotic performance guarantees in settings with independent training samples. However, these guarantees are not…

Optimization and Control · Mathematics 2021-12-13 Yafei Wang , Bo Pan , Wei Tu , Peng Liu , Bei Jiang , Chao Gao , Wei Lu , Shangling Jui , Linglong Kong

The simplest and most widely applied method for guaranteeing differential privacy is to add instance-independent noise to a statistic of interest that is scaled to its global sensitivity. However, global sensitivity is a worst-case notion…

Statistics Theory · Mathematics 2019-06-10 Mark Bun , Thomas Steinke

Standard bandit algorithms that assume continual reallocation of measurement effort are challenging to implement due to delayed feedback and infrastructural/organizational difficulties. Motivated by practical instances involving a handful…

Machine Learning · Computer Science 2023-08-16 Ethan Che , Hongseok Namkoong

The notion of replicable algorithms was introduced in Impagliazzo et al. [STOC '22] to describe randomized algorithms that are stable under the resampling of their inputs. More precisely, a replicable algorithm gives the same output with…

Machine Learning · Computer Science 2023-03-28 Mark Bun , Marco Gaboardi , Max Hopkins , Russell Impagliazzo , Rex Lei , Toniann Pitassi , Satchit Sivakumar , Jessica Sorrell

In this paper, we introduce a notion of algorithmic stability called typical stability. When our goal is to release real-valued queries (statistics) computed over a dataset, this notion does not require the queries to be of bounded…

Machine Learning · Computer Science 2016-09-20 Raef Bassily , Yoav Freund

We propose, implement, and evaluate a new algorithm for releasing answers to very large numbers of statistical queries like $k$-way marginals, subject to differential privacy. Our algorithm makes adaptive use of a continuous relaxation of…

Machine Learning · Computer Science 2021-06-24 Sergul Aydore , William Brown , Michael Kearns , Krishnaram Kenthapadi , Luca Melis , Aaron Roth , Ankit Siva

We consider the problem of hypothesis testing for discrete distributions. In the standard model, where we have sample access to an underlying distribution $p$, extensive research has established optimal bounds for uniformity testing,…

Machine Learning · Computer Science 2024-12-03 Maryam Aliakbarpour , Piotr Indyk , Ronitt Rubinfeld , Sandeep Silwal

Reuse of data in adaptive workflows poses challenges regarding overfitting and the statistical validity of results. Previous work has demonstrated that interacting with data via differentially private algorithms can mitigate overfitting,…

Machine Learning · Computer Science 2025-11-13 Neil G. Marchant , Benjamin I. P. Rubinstein

We present \textit{universal} estimators for the statistical mean, variance, and scale (in particular, the interquartile range) under pure differential privacy. These estimators are universal in the sense that they work on an arbitrary,…

Cryptography and Security · Computer Science 2023-04-04 Wei Dong , Ke Yi

We study the problem of estimating finite sample confidence intervals of the mean of a normal population under the constraint of differential privacy. We consider both the known and unknown variance cases and construct differentially…

Cryptography and Security · Computer Science 2017-11-13 Vishesh Karwa , Salil Vadhan
‹ Prev 1 2 3 10 Next ›