English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Class incremental learning (CIL) aims to recognize both the old and newclasses along the increment tasks. Deep neural networks in CIL suffer fromcatastrophic forgetting and some approaches rely on saving exemplars fromprevious tasks, known as the exemplar-based setting, to alleviate this problem.On the contrary, this paper focuses on the Exemplar-Free setting with no oldclass sample preserved. Balancing the plasticity and stability in deep featurelearning with only supervision from new classes is more challenging. Mostexisting Exemplar-Free CIL methods report the overall performance only and lackfurther analysis. In this work, different methods are examined withcomplementary metrics in greater detail. Moreover, we propose a simple CILmethod, Rotation Augmented Distillation (RAD), which achieves one of thetop-tier performances under the Exemplar-Free setting. Detailed analysis showsour RAD benefits from the superior balance between plasticity and stability.Finally, more challenging exemplar-free settings with fewer initial classes areundertaken for further demonstrations and comparisons among thestate-of-the-art methods.

Rotation Augmented Distillation for Exemplar-Free Class Incremental Learning with Detailed Analysis