Adaptive task sampling for meta-learning
Meta-learning methods have been extensively studied and applied in computer vision, especially for few-shot classification tasks. The key idea of meta-learning for few-shot classification is to mimic the few-shot situations faced at test time by randomly sampling classes in meta-training data to con...
Saved in:
Main Authors: | , , , , , |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2020
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/5293 https://ink.library.smu.edu.sg/context/sis_research/article/6296/viewcontent/ECCV20_GCP.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Singapore Management University |
Language: | English |
id |
sg-smu-ink.sis_research-6296 |
---|---|
record_format |
dspace |
spelling |
sg-smu-ink.sis_research-62962020-09-09T04:46:12Z Adaptive task sampling for meta-learning LIU, Chenghao WANG, Zhihao SAHOO, Doyen FANG, Yuan ZHANG, Kun HOI, Steven C. H. Meta-learning methods have been extensively studied and applied in computer vision, especially for few-shot classification tasks. The key idea of meta-learning for few-shot classification is to mimic the few-shot situations faced at test time by randomly sampling classes in meta-training data to construct fewshot tasks for episodic training. While a rich line of work focuses solely on how to extract meta-knowledge across tasks, we exploit the complementary problem on how to generate informative tasks. We argue that the randomly sampled tasks could be sub-optimal and uninformative (e.g., the task of classifying “dog” from “laptop” is often trivial) to the meta-learner. In this paper, we propose an adaptive task sampling method to improve the generalization performance. Unlike instance based sampling, task based sampling is much more challenging due to the implicit definition of the task in each episode. Therefore, we accordingly propose a greedy class-pair based sampling method, which selects difficult tasks according to class-pair potentials. We evaluate our adaptive task sampling method on two few-shot classification benchmarks, and it achieves consistent improvements across different feature backbones, meta-learning algorithms and datasets. 2020-08-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/5293 https://ink.library.smu.edu.sg/context/sis_research/article/6296/viewcontent/ECCV20_GCP.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Meta-learning task sampling few-shot learning Databases and Information Systems Theory and Algorithms |
institution |
Singapore Management University |
building |
SMU Libraries |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
SMU Libraries |
collection |
InK@SMU |
language |
English |
topic |
Meta-learning task sampling few-shot learning Databases and Information Systems Theory and Algorithms |
spellingShingle |
Meta-learning task sampling few-shot learning Databases and Information Systems Theory and Algorithms LIU, Chenghao WANG, Zhihao SAHOO, Doyen FANG, Yuan ZHANG, Kun HOI, Steven C. H. Adaptive task sampling for meta-learning |
description |
Meta-learning methods have been extensively studied and applied in computer vision, especially for few-shot classification tasks. The key idea of meta-learning for few-shot classification is to mimic the few-shot situations faced at test time by randomly sampling classes in meta-training data to construct fewshot tasks for episodic training. While a rich line of work focuses solely on how to extract meta-knowledge across tasks, we exploit the complementary problem on how to generate informative tasks. We argue that the randomly sampled tasks could be sub-optimal and uninformative (e.g., the task of classifying “dog” from “laptop” is often trivial) to the meta-learner. In this paper, we propose an adaptive task sampling method to improve the generalization performance. Unlike instance based sampling, task based sampling is much more challenging due to the implicit definition of the task in each episode. Therefore, we accordingly propose a greedy class-pair based sampling method, which selects difficult tasks according to class-pair potentials. We evaluate our adaptive task sampling method on two few-shot classification benchmarks, and it achieves consistent improvements across different feature backbones, meta-learning algorithms and datasets. |
format |
text |
author |
LIU, Chenghao WANG, Zhihao SAHOO, Doyen FANG, Yuan ZHANG, Kun HOI, Steven C. H. |
author_facet |
LIU, Chenghao WANG, Zhihao SAHOO, Doyen FANG, Yuan ZHANG, Kun HOI, Steven C. H. |
author_sort |
LIU, Chenghao |
title |
Adaptive task sampling for meta-learning |
title_short |
Adaptive task sampling for meta-learning |
title_full |
Adaptive task sampling for meta-learning |
title_fullStr |
Adaptive task sampling for meta-learning |
title_full_unstemmed |
Adaptive task sampling for meta-learning |
title_sort |
adaptive task sampling for meta-learning |
publisher |
Institutional Knowledge at Singapore Management University |
publishDate |
2020 |
url |
https://ink.library.smu.edu.sg/sis_research/5293 https://ink.library.smu.edu.sg/context/sis_research/article/6296/viewcontent/ECCV20_GCP.pdf |
_version_ |
1770575373869776896 |