Few-Shot Class Incremental Learning with Generative Feature Replay

Abhilash Shankarampeta, Koichiro Yamauchi

Abstract

The humans can learn novel concepts from only a few examples effortlessly and learn additional tasks without forgetting previous ones. Making machines to learn incrementally from only a few instances is very challenging due to catastrophic forgetting between new and previously learned tasks; this can be solved by generative image replay. However, image generation with only a few examples is a challenging task. In this work, we propose a feature replay approach instead of image replay for few-shot learning scenarios. A feature extractor with feature distillation is combined with feature replay at the classifier level to tackle catastrophic forgetting.

Download


Paper Citation


in Harvard Style

Shankarampeta A. and Yamauchi K. (2021). Few-Shot Class Incremental Learning with Generative Feature Replay.In Proceedings of the 10th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-486-2, pages 259-267. DOI: 10.5220/0010246602590267


in Bibtex Style

@conference{icpram21,
author={Abhilash Shankarampeta and Koichiro Yamauchi},
title={Few-Shot Class Incremental Learning with Generative Feature Replay},
booktitle={Proceedings of the 10th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2021},
pages={259-267},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010246602590267},
isbn={978-989-758-486-2},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 10th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Few-Shot Class Incremental Learning with Generative Feature Replay
SN - 978-989-758-486-2
AU - Shankarampeta A.
AU - Yamauchi K.
PY - 2021
SP - 259
EP - 267
DO - 10.5220/0010246602590267