The Hitchhiker's Guide to Prior-Shift Adaptation

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F22%3A00363002" target="_blank" >RIV/68407700:21230/22:00363002 - isvavai.cz</a>

Result on the web

<a href="https://doi.org/10.1109/WACV51458.2022.00209" target="_blank" >https://doi.org/10.1109/WACV51458.2022.00209</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/WACV51458.2022.00209" target="_blank" >10.1109/WACV51458.2022.00209</a>

Result language

angličtina

Original language name

The Hitchhiker's Guide to Prior-Shift Adaptation

Original language description

In many computer vision classification tasks, class priors at test time often differ from priors on the training set. In the case of such prior shift, classifiers must be adapted correspondingly to maintain close to optimal performance. This paper analyzes methods for adaptation of probabilistic classifiers to new priors and for estimating new priors on an unlabeled test set. We propose a novel method to address a known issue of prior estimation methods based on confusion matrices, where inconsistent estimates of decision probabilities and confusion matrices lead to negative values in the estimated priors. Experiments on fine-grained image classification datasets provide insight into the best practice of prior shift estimation and classifier adaptation, and show that the proposed method achieves state-of-the-art results in prior adaptation. Applying the best practice to two tasks with naturally imbalanced priors, learning from web-crawled images and plant species classification, increased the recognition accuracy by 1.1% and 3.4% respectively.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

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

Project

<a href="/en/project/EF16_019%2F0000765" target="_blank" >EF16_019/0000765: Research Center for Informatics</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Publication year

2022

Confidentiality

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

Article name in the collection

Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2022

ISBN

978-1-6654-0915-5

ISSN

2472-6737

e-ISSN

2642-9381

Number of pages

9

Pages from-to

2031-2039

Publisher name

IEEE Computer Society

Place of publication

USA

Event location

Waikoloa

Event date

Jan 3, 2022

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

000800471202010