A Baseline for Few-Shot Image Classification
Guneet Singh Dhillon, Pratik Chaudhari, Avinash Ravichandran, Stefano Soatto
Abstract
Fine-tuning a deep network trained with the standard cross-entropy loss is a strong baseline for few-shot learning. When fine-tuned transductively, this outperforms the current state-of-the-art on standard datasets such as Mini-ImageNet, Tiered-ImageNet, CIFAR-FS and FC-100 with the same hyper-parameters. The simplicity of this approach enables us to demonstrate the first few-shot learning results on the ImageNet-21k dataset. We find that using a large number of meta-training classes results in high few-shot accuracies even for a large number of few-shot classes. We do not advocate our approach as the solution for few-shot learning, but simply use the results to highlight limitations of current benchmarks and few-shot protocols. We perform extensive studies on benchmark datasets to propose a metric that quantifies the "hardness" of a few-shot episode. This metric can be used to report the performance of few-shot algorithms in a more systematic way.
BibTeX
@inproceedings{
Dhillon2020A,
title={A Baseline for Few-Shot Image Classification},
author={Guneet Singh Dhillon and Pratik Chaudhari and Avinash Ravichandran and Stefano Soatto},
booktitle={International Conference on Learning Representations},
year={2020},
url={https://openreview.net/forum?id=rylXBkrYDS}
}