SCOD: From Heuristics to Theory

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F25%3A00378258" target="_blank" >RIV/68407700:21230/25:00378258 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1007/978-3-031-72907-2_25" target="_blank" >https://doi.org/10.1007/978-3-031-72907-2_25</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-72907-2_25" target="_blank" >10.1007/978-3-031-72907-2_25</a>

Jazyk výsledku

angličtina

Název v původním jazyce

SCOD: From Heuristics to Theory

Popis výsledku v původním jazyce

This paper addresses the problem of designing reliable prediction models that abstain from predictions when faced with uncertain or out-of-distribution samples - a recently proposed problem known as Selective Classification in the presence of Out-of-Distribution data (SCOD). We make three key contributions to SCOD. Firstly, we demonstrate that the optimal SCOD strategy involves a Bayes classifier for in-distribution (ID) data and a selector represented as a stochastic linear classifier in a 2D space, using i) the conditional risk of the ID classifier, and ii) the likelihood ratio of ID and out-of-distribution (OOD) data as input. This contrasts with suboptimal strategies from current OOD detection methods and the Softmax Information Retaining Combination (SIRC), specifically developed for SCOD. Secondly, we establish that in a distribution-free setting, the SCOD problem is not Probably Approximately Correct learnable when relying solely on an ID data sample. Third, we introduce POSCOD, a simple method for learning a plugin estimate of the optimal SCOD strategy from both an ID data sample and an unlabeled mixture of ID and OOD data. Our empirical results confirm the theoretical findings and demonstrate that our proposed method, POSCOD, outperforms existing OOD methods in effectively addressing the SCOD problem.

Název v anglickém jazyce

SCOD: From Heuristics to Theory

Popis výsledku anglicky

This paper addresses the problem of designing reliable prediction models that abstain from predictions when faced with uncertain or out-of-distribution samples - a recently proposed problem known as Selective Classification in the presence of Out-of-Distribution data (SCOD). We make three key contributions to SCOD. Firstly, we demonstrate that the optimal SCOD strategy involves a Bayes classifier for in-distribution (ID) data and a selector represented as a stochastic linear classifier in a 2D space, using i) the conditional risk of the ID classifier, and ii) the likelihood ratio of ID and out-of-distribution (OOD) data as input. This contrasts with suboptimal strategies from current OOD detection methods and the Softmax Information Retaining Combination (SIRC), specifically developed for SCOD. Secondly, we establish that in a distribution-free setting, the SCOD problem is not Probably Approximately Correct learnable when relying solely on an ID data sample. Third, we introduce POSCOD, a simple method for learning a plugin estimate of the optimal SCOD strategy from both an ID data sample and an unlabeled mixture of ID and OOD data. Our empirical results confirm the theoretical findings and demonstrate that our proposed method, POSCOD, outperforms existing OOD methods in effectively addressing the SCOD problem.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Computer Vision – ECCV 2024, Part LXXXIV

ISBN

978-3-031-72906-5

ISSN

0302-9743

e-ISSN

1611-3349

Počet stran výsledku

18

Strana od-do

424-441

Název nakladatele

Springer, Cham

Místo vydání

—

Místo konání akce

Milano

Datum konání akce

29. 9. 2024

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

001352890300025