Continual Novelty Detection

Rahaf Aljundi, Daniel Olmeda Reino, Nikolay Chumerin, Richard E. Turner
Proceedings of The 1st Conference on Lifelong Learning Agents, PMLR 199:1004-1025, 2022.

Abstract

Novelty Detection methods identify samples that are not representative of a model’s training set thereby flagging misleading predictions and bringing a greater flexibility and transparency at deployment time. However, research in this area has only considered Novelty Detection in the offline setting. Recently, there has been a growing realization in the computer vision community that applications demand a more flexible framework - Continual Learning - where new batches of data representing new domains, new classes or new tasks become available at different points in time. In this setting, Novelty Detection becomes more important, interesting and challenging. This work identifies the crucial link between the two problems and investigates the Novelty Detection problem under the Continual Learning setting. We formulate the Continual Novelty Detection problem and present a benchmark, where we compare several Novelty Detection methods under different Continual Learning settings. We show that Continual Learning affects the behaviour of novelty detection algorithms, while novelty detection can pinpoint insights in the behaviour of a continual learner. We further propose baselines and discuss possible research directions. We believe that the coupling of the two problems is a promising direction to bring vision models into practice.

Cite this Paper


BibTeX
@InProceedings{pmlr-v199-aljundi22a, title = {Continual Novelty Detection}, author = {Aljundi, Rahaf and Reino, Daniel Olmeda and Chumerin, Nikolay and Turner, Richard E.}, booktitle = {Proceedings of The 1st Conference on Lifelong Learning Agents}, pages = {1004--1025}, year = {2022}, editor = {Chandar, Sarath and Pascanu, Razvan and Precup, Doina}, volume = {199}, series = {Proceedings of Machine Learning Research}, month = {22--24 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v199/aljundi22a/aljundi22a.pdf}, url = {https://proceedings.mlr.press/v199/aljundi22a.html}, abstract = {Novelty Detection methods identify samples that are not representative of a model’s training set thereby flagging misleading predictions and bringing a greater flexibility and transparency at deployment time. However, research in this area has only considered Novelty Detection in the offline setting. Recently, there has been a growing realization in the computer vision community that applications demand a more flexible framework - Continual Learning - where new batches of data representing new domains, new classes or new tasks become available at different points in time. In this setting, Novelty Detection becomes more important, interesting and challenging. This work identifies the crucial link between the two problems and investigates the Novelty Detection problem under the Continual Learning setting. We formulate the Continual Novelty Detection problem and present a benchmark, where we compare several Novelty Detection methods under different Continual Learning settings. We show that Continual Learning affects the behaviour of novelty detection algorithms, while novelty detection can pinpoint insights in the behaviour of a continual learner. We further propose baselines and discuss possible research directions. We believe that the coupling of the two problems is a promising direction to bring vision models into practice.} }
Endnote
%0 Conference Paper %T Continual Novelty Detection %A Rahaf Aljundi %A Daniel Olmeda Reino %A Nikolay Chumerin %A Richard E. Turner %B Proceedings of The 1st Conference on Lifelong Learning Agents %C Proceedings of Machine Learning Research %D 2022 %E Sarath Chandar %E Razvan Pascanu %E Doina Precup %F pmlr-v199-aljundi22a %I PMLR %P 1004--1025 %U https://proceedings.mlr.press/v199/aljundi22a.html %V 199 %X Novelty Detection methods identify samples that are not representative of a model’s training set thereby flagging misleading predictions and bringing a greater flexibility and transparency at deployment time. However, research in this area has only considered Novelty Detection in the offline setting. Recently, there has been a growing realization in the computer vision community that applications demand a more flexible framework - Continual Learning - where new batches of data representing new domains, new classes or new tasks become available at different points in time. In this setting, Novelty Detection becomes more important, interesting and challenging. This work identifies the crucial link between the two problems and investigates the Novelty Detection problem under the Continual Learning setting. We formulate the Continual Novelty Detection problem and present a benchmark, where we compare several Novelty Detection methods under different Continual Learning settings. We show that Continual Learning affects the behaviour of novelty detection algorithms, while novelty detection can pinpoint insights in the behaviour of a continual learner. We further propose baselines and discuss possible research directions. We believe that the coupling of the two problems is a promising direction to bring vision models into practice.
APA
Aljundi, R., Reino, D.O., Chumerin, N. & Turner, R.E.. (2022). Continual Novelty Detection. Proceedings of The 1st Conference on Lifelong Learning Agents, in Proceedings of Machine Learning Research 199:1004-1025 Available from https://proceedings.mlr.press/v199/aljundi22a.html.

Related Material