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

This paper considers incremental few-shot learning, which requires a model tocontinually recognize new categories with only a few examples provided. Ourstudy shows that existing methods severely suffer from catastrophic forgetting,a well-known problem in incremental learning, which is aggravated due to datascarcity and imbalance in the few-shot setting. Our analysis further suggeststhat to prevent catastrophic forgetting, actions need to be taken in theprimitive stage -- the training of base classes instead of later few-shotlearning sessions. Therefore, we propose to search for flat local minima of thebase training objective function and then fine-tune the model parameters withinthe flat region on new tasks. In this way, the model can efficiently learn newclasses while preserving the old ones. Comprehensive experimental resultsdemonstrate that our approach outperforms all prior state-of-the-art methodsand is very close to the approximate upper bound. The source code is availableat https://github.com/moukamisama/F2M.

Overcoming Catastrophic Forgetting in Incremental Few-Shot Learning by Finding Flat Minima