A Strong Baseline for Semi-Supervised Incremental Few-Shot Learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhao, Linglan, Guo, Dashan, Xu, Yunlu, Qiao, Liang, Cheng, Zhanzhan, Pu, Shiliang, Niu, Yi, Fang, Xiangzhong
Natura: Preprint
Pubblicazione: 2021
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913567365660672
author Zhao, Linglan
Guo, Dashan
Xu, Yunlu
Qiao, Liang
Cheng, Zhanzhan
Pu, Shiliang
Niu, Yi
Fang, Xiangzhong
author_facet Zhao, Linglan
Guo, Dashan
Xu, Yunlu
Qiao, Liang
Cheng, Zhanzhan
Pu, Shiliang
Niu, Yi
Fang, Xiangzhong
contents Few-shot learning (FSL) aims to learn models that generalize to novel classes with limited training samples. Recent works advance FSL towards a scenario where unlabeled examples are also available and propose semi-supervised FSL methods. Another line of methods also cares about the performance of base classes in addition to the novel ones and thus establishes the incremental FSL scenario. In this paper, we generalize the above two under a more realistic yet complex setting, named by Semi-Supervised Incremental Few-Shot Learning (S2 I-FSL). To tackle the task, we propose a novel paradigm containing two parts: (1) a well-designed meta-training algorithm for mitigating ambiguity between base and novel classes caused by unreliable pseudo labels and (2) a model adaptation mechanism to learn discriminative features for novel classes while preserving base knowledge using few labeled and all the unlabeled data. Extensive experiments on standard FSL, semi-supervised FSL, incremental FSL, and the firstly built S2 I-FSL benchmarks demonstrate the effectiveness of our proposed method.
format Preprint
id arxiv_https___arxiv_org_abs_2110_11128
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle A Strong Baseline for Semi-Supervised Incremental Few-Shot Learning
Zhao, Linglan
Guo, Dashan
Xu, Yunlu
Qiao, Liang
Cheng, Zhanzhan
Pu, Shiliang
Niu, Yi
Fang, Xiangzhong
Computer Vision and Pattern Recognition
Few-shot learning (FSL) aims to learn models that generalize to novel classes with limited training samples. Recent works advance FSL towards a scenario where unlabeled examples are also available and propose semi-supervised FSL methods. Another line of methods also cares about the performance of base classes in addition to the novel ones and thus establishes the incremental FSL scenario. In this paper, we generalize the above two under a more realistic yet complex setting, named by Semi-Supervised Incremental Few-Shot Learning (S2 I-FSL). To tackle the task, we propose a novel paradigm containing two parts: (1) a well-designed meta-training algorithm for mitigating ambiguity between base and novel classes caused by unreliable pseudo labels and (2) a model adaptation mechanism to learn discriminative features for novel classes while preserving base knowledge using few labeled and all the unlabeled data. Extensive experiments on standard FSL, semi-supervised FSL, incremental FSL, and the firstly built S2 I-FSL benchmarks demonstrate the effectiveness of our proposed method.
title A Strong Baseline for Semi-Supervised Incremental Few-Shot Learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2110.11128