Experiments in Cross-domain Few-shot Learning for Image Classification: Extended Abstract

Hongyu Wang, Huon Fraser, Henry Gouk, Eibe Frank, Bernhard Pfahringer, Michael Mayo, Geoff Holmes
ECMLPKDD Workshop on Meta-Knowledge Transfer, PMLR 191:81-83, 2022.

Abstract

We summarise experiments evaluating cross-domain few-shot learning (CDFSL) with feature extractors trained on ImageNet. The work explores the transfer performance of extracted features on five target domains with different degrees of similarity to ImageNet. These experiments compare robust classifiers and normalisation methods, consider multi-instance learning algorithms, and evaluate the effect of using features extracted by different ResNet backbones at various levels of their convolutional hierarchies. The cosine similarity classifier and 1-vs-rest logistic regression with $\ell_2$ regularisation are the top-performing robust classifiers in the evaluation, and $\ell_2$ normalisation improves performance on all five target domains when using LDA as the robust classifier. The results also show that feature extractors with the highest capacity do not always achieve the best CDFSL performance. Lastly, simple multi-instance learning methods are shown to improve classifier accuracy.

Cite this Paper

BibTeX
@InProceedings{pmlr-v191-wang22a, title = {Experiments in Cross-domain Few-shot Learning for Image Classification: Extended Abstract}, author = {Wang, Hongyu and Fraser, Huon and Gouk, Henry and Frank, Eibe and Pfahringer, Bernhard and Mayo, Michael and Holmes, Geoff}, booktitle = {ECMLPKDD Workshop on Meta-Knowledge Transfer}, pages = {81--83}, year = {2022}, editor = {Brazdil, Pavel and van Rijn, Jan N. and Gouk, Henry and Mohr, Felix}, volume = {191}, series = {Proceedings of Machine Learning Research}, month = {23 Sep}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v191/wang22a/wang22a.pdf}, url = {https://proceedings.mlr.press/v191/wang22a.html}, abstract = {We summarise experiments evaluating cross-domain few-shot learning (CDFSL) with feature extractors trained on ImageNet. The work explores the transfer performance of extracted features on five target domains with different degrees of similarity to ImageNet. These experiments compare robust classifiers and normalisation methods, consider multi-instance learning algorithms, and evaluate the effect of using features extracted by different ResNet backbones at various levels of their convolutional hierarchies. The cosine similarity classifier and 1-vs-rest logistic regression with $\ell_2$ regularisation are the top-performing robust classifiers in the evaluation, and $\ell_2$ normalisation improves performance on all five target domains when using LDA as the robust classifier. The results also show that feature extractors with the highest capacity do not always achieve the best CDFSL performance. Lastly, simple multi-instance learning methods are shown to improve classifier accuracy.} }
Endnote
%0 Conference Paper %T Experiments in Cross-domain Few-shot Learning for Image Classification: Extended Abstract %A Hongyu Wang %A Huon Fraser %A Henry Gouk %A Eibe Frank %A Bernhard Pfahringer %A Michael Mayo %A Geoff Holmes %B ECMLPKDD Workshop on Meta-Knowledge Transfer %C Proceedings of Machine Learning Research %D 2022 %E Pavel Brazdil %E Jan N. van Rijn %E Henry Gouk %E Felix Mohr %F pmlr-v191-wang22a %I PMLR %P 81--83 %U https://proceedings.mlr.press/v191/wang22a.html %V 191 %X We summarise experiments evaluating cross-domain few-shot learning (CDFSL) with feature extractors trained on ImageNet. The work explores the transfer performance of extracted features on five target domains with different degrees of similarity to ImageNet. These experiments compare robust classifiers and normalisation methods, consider multi-instance learning algorithms, and evaluate the effect of using features extracted by different ResNet backbones at various levels of their convolutional hierarchies. The cosine similarity classifier and 1-vs-rest logistic regression with $\ell_2$ regularisation are the top-performing robust classifiers in the evaluation, and $\ell_2$ normalisation improves performance on all five target domains when using LDA as the robust classifier. The results also show that feature extractors with the highest capacity do not always achieve the best CDFSL performance. Lastly, simple multi-instance learning methods are shown to improve classifier accuracy.
APA
Wang, H., Fraser, H., Gouk, H., Frank, E., Pfahringer, B., Mayo, M. & Holmes, G.. (2022). Experiments in Cross-domain Few-shot Learning for Image Classification: Extended Abstract. ECMLPKDD Workshop on Meta-Knowledge Transfer, in Proceedings of Machine Learning Research 191:81-83 Available from https://proceedings.mlr.press/v191/wang22a.html.

Related Material