Fetching the paper…

Reading the bibliography…

Bayesian Adaptive Data Analysis Guarantees from Subgaussianity · Around

2016

Bayesian Adaptive Data Analysis Guarantees from Subgaussianity

Elder, Sam

Original
PDF
Post on X

Understand

The new field of adaptive data analysis seeks to provide algorithms and provable guarantees for models of machine learning that allow researchers to reuse their data, which normally falls outside of the usual statistical paradigm of static data analysis.

In 2014, Dwork, Feldman, Hardt, Pitassi, Reingold and Roth introduced one potential model and proposed several solutions based on differential privacy.
In previous work in 2016, we described a problem with this model and instead proposed a Bayesian variant, but also found that the analogous Bayesian methods cannot achieve the same statistical guarantees as in the static case.
In this paper, we prove the first positive results for the Bayesian model, showing that with a Dirichlet prior, the posterior mean algorithm indeed matches the statistical guarantees of the static case.

Built on

A note on screening regression equations
David A Freedman · 1983
Earlier work this paper cites.
Metric characterization of random variables and random processes
Valeriĭ Vladimirovich Buldygin and I͡U V Kozachenko · 2000
Earlier work this paper cites.
Chi distribution
Eric W Weisstein · 2003
Earlier work this paper cites.
Handbook of beta distribution and its applications
Arjun K Gupta and Saralees Nadarajah · 2004
Earlier work this paper cites.

Similar

The sub-gaussian norm of a binary random variable
V Buldygin and K Moskvichova · 2013
Cited alongside, same era.
Probability in high dimension
Ramon van Handel · 2014
Cited alongside, same era.
The ladder: A reliable leaderboard for machine learning competitions
Original
Avrim Blum and Moritz Hardt · 2015
Cited alongside, same era.

Then

Preserving statistical validity in adaptive data analysis
Cynthia Dwork, Vitaly Feldman, Moritz Hardt, Toniann Pitassi, Omer Reingold, and Aaron Leon Roth · 2015
Later among the works it cites.
Challenges in bayesian adaptive data analysis
Original
Sam Elder · 2016
Closest in time.
Statistical limits of spiked tensor models
Amelia Perry, Alexander S. Wein, and Afonso S. Bandeira · 2016
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…