File(s) under embargo

Reason: Publisher requirement. Embargo date will be amended following publication.

9

year(s)

11

month(s)

4

day(s)

until file(s) become available

Open set recognition through Monte Carlo dropout-based uncertainty

journal contribution
posted on 17.03.2021, 10:50 by Xiaojie Yin, Qinghua Hu, Gerald Schaefer
Open set recognition has received much attention in recent years. In this paper, we present a novel open set recognition method that is able to obtain improved recognition by applying Monte Carlo Dropout to capture uncertainty in order to yield high quality predicted probabilities. Experimental results on six benchmark datasets show that our method gives better open set recognition performance than other stateof-the-art methods, with at least 6.4%, 3.9%, 2.9% and 1.0% performance increase in AUROC on the challenging object datasets CIFAR-10, CIFAR+10, CIFAR+50 and TinyImageNet respectively. We also perform an analysis on the benefits of combining predictive uncertainty with an EVT-based open set recognition model which indicates that Monte Carlo Dropout-based uncertainty allows to obtain high quality predicted probabilities and to learn more accurate open set recognition scores. This, in turn, helps to reduce the overlap between known and unknown classes, thus making them more separable.

Funding

National Natural Science Foundation of China under Grants 61925602 and 61732011

History

School

  • Science

Department

  • Computer Science

Published in

International Journal of Bio-Inspired Computation

Publisher

Inderscience

Version

AM (Accepted Manuscript)

Rights holder

© Inderscience

Publisher statement

This paper was accepted for publication in the journal International Journal of Bio-Inspired Computation and the definitive published version is available at [insert DOI link].

Acceptance date

22/02/2021

ISSN

1758-0366

eISSN

1758-0374

Language

en

Depositor

Dr Gerald Schaefer. Deposit date: 16 March 2021

Exports

Loughborough Publications

Categories

Exports