Title :

KSD Aggregated Goodness-of-fit Test

Author(s) :

Schrab, Antonin [Auteur]
MOdel for Data Analysis and Learning [MODAL]
The Inria London Programme [Inria-London]
Department of Computer science [University College of London] [UCL-CS]
Gatsby Computational Neuroscience Unit
Guedj, Benjamin [Auteur]
MOdel for Data Analysis and Learning [MODAL]
The Inria London Programme [Inria-London]
The Alan Turing Institute
Department of Computer science [University College of London] [UCL-CS]
Gretton, Arthur [Auteur]
Gatsby Computational Neuroscience Unit

Conference title :

Conference on Neural Information Processing Systems

City :

New Orleans

Country :

Etats-Unis d'Amérique

Start date of the conference :

2022-11-28

Book title :

PMLR

HAL domain(s) :

Statistiques [stat]/Machine Learning [stat.ML]
Informatique [cs]/Apprentissage [cs.LG]
Statistiques [stat]/Théorie [stat.TH]

English abstract : [en]

We investigate properties of goodness-of-fit tests based on the Kernel Stein Discrepancy (KSD). We introduce a strategy to construct a test, called KSDAGG, which aggregates multiple tests with different kernels. KSDAGG ...
Show more >We investigate properties of goodness-of-fit tests based on the Kernel Stein Discrepancy (KSD). We introduce a strategy to construct a test, called KSDAGG, which aggregates multiple tests with different kernels. KSDAGG avoids splitting the data to perform kernel selection (which leads to a loss in test power), and rather maximises the test power over a collection of kernels. We provide theoretical guarantees on the power of KSDAGG: we show it achieves the smallest uniform separation rate of the collection, up to a logarithmic term. KSDAGG can be computed exactly in practice as it relies either on a parametric bootstrap or on a wild bootstrap to estimate the quantiles and the level corrections. In particular, for the crucial choice of bandwidth of a fixed kernel, it avoids resorting to arbitrary heuristics (such as median or standard deviation) or to data splitting. We find on both synthetic and real-world data that KSDAGG outperforms other state-of-the-art adaptive KSDbased goodness-of-fit testing procedures.Show less >

Language :

Anglais

Peer reviewed article :

Oui

Audience :

Internationale

Popular science :

Non

Comment :

9 pages

Collections :

• Laboratoire Paul Painlevé - UMR 8524

Source :

Harvested from HAL