Recall@k Surrogate Loss with Large Batches and Similarity Mixup

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F22%3A00362948" target="_blank" >RIV/68407700:21230/22:00362948 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1109/CVPR52688.2022.00735" target="_blank" >https://doi.org/10.1109/CVPR52688.2022.00735</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/CVPR52688.2022.00735" target="_blank" >10.1109/CVPR52688.2022.00735</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Recall@k Surrogate Loss with Large Batches and Similarity Mixup

Popis výsledku v původním jazyce

This work focuses on learning deep visual representation models for retrieval by exploring the interplay between a new loss function, the batch size, and a new regularization approach. Direct optimization, by gradient descent, of an evaluation metric, is not possible when it is non-differentiable, which is the case for recall in retrieval. A differentiable surrogate loss for the recall is proposed in this work. Using an implementation that sidesteps the hardware constraints of the GPU memory, the method trains with a very large batch size, which is essential for metrics computed on the entire retrieval database. It is assisted by an efficient mixup regularization approach that operates on pairwise scalar similarities and virtually increases the batch size further. The suggested method achieves state-of-the-art performance in several image retrieval benchmarks when used for deep metric learning. For instance-level recognition, the method outperforms similar approaches that train using an approximation of average precision.

Název v anglickém jazyce

Recall@k Surrogate Loss with Large Batches and Similarity Mixup

Popis výsledku anglicky

This work focuses on learning deep visual representation models for retrieval by exploring the interplay between a new loss function, the batch size, and a new regularization approach. Direct optimization, by gradient descent, of an evaluation metric, is not possible when it is non-differentiable, which is the case for recall in retrieval. A differentiable surrogate loss for the recall is proposed in this work. Using an implementation that sidesteps the hardware constraints of the GPU memory, the method trains with a very large batch size, which is essential for metrics computed on the entire retrieval database. It is assisted by an efficient mixup regularization approach that operates on pairwise scalar similarities and virtually increases the batch size further. The suggested method achieves state-of-the-art performance in several image retrieval benchmarks when used for deep metric learning. For instance-level recognition, the method outperforms similar approaches that train using an approximation of average precision.

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

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2022

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceeding 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

ISBN

978-1-6654-6946-3

ISSN

1063-6919

e-ISSN

2575-7075

Počet stran výsledku

10

Strana od-do

7492-7501

Název nakladatele

IEEE

Místo vydání

Piscataway

Místo konání akce

New Orleans, Louisiana

Datum konání akce

19. 6. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000870759100033