Re-assessing Accuracy Degradation: A Framework for Understanding DNN Behavior on Similar-but-not-identical Test Datasets

Unpublished conference/Abstract (Scientific congresses and symposiums)

Anzaku, Esla Timothy; Wang, Haohan; Babalola, Ajiboye et al.

2024 • The 16th Asian Conference on Machine Learning (ACML) 2024

Permalink
https://hdl.handle.net/2268/328675

acml_2024_eta_final.pdf

Author postprint (2.55 MB)

All documents in ORBi are protected by a user license.

Disciplines :

Computer science

Author, co-author :

Anzaku, Esla Timothy

Wang, Haohan

Babalola, Ajiboye

Van Messem, Arnout ; Université de Liège - ULiège > Mathematics

De Neve, Wesley

Language :

English

Title :

Re-assessing Accuracy Degradation: A Framework for Understanding DNN Behavior on Similar-but-not-identical Test Datasets

Publication date :

December 2024

Event name :

The 16th Asian Conference on Machine Learning (ACML) 2024

Event place :

Hanoi, Vietnam

Event date :

December 5-8, 2024

Audience :

International

Available on ORBi :

since 24 February 2025

Number of views

28 (0 by ULiège)

Number of downloads

16 (0 by ULiège)