TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK	REMOVE
Few-Shot Image Classification	Meta-Dataset	URT	Accuracy	72.15	# 7
Few-Shot Image Classification	Meta-Dataset Rank	URT	Mean Rank	2.85	# 1

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/a-universal-representation-transformer-layer/few-shot-image-classification-on-meta-dataset-1)](https://paperswithcode.com/sota/few-shot-image-classification-on-meta-dataset-1?p=a-universal-representation-transformer-layer)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/a-universal-representation-transformer-layer/few-shot-image-classification-on-meta-dataset)](https://paperswithcode.com/sota/few-shot-image-classification-on-meta-dataset?p=a-universal-representation-transformer-layer)`

A Universal Representation Transformer Layer for Few-Shot Image Classification

ICLR 2021 · Lu Liu, William Hamilton, Guodong Long, Jing Jiang, Hugo Larochelle ·

Few-shot classification aims to recognize unseen classes when presented with only a small number of samples. We consider the problem of multi-domain few-shot image classification, where unseen classes and examples come from diverse data sources. This problem has seen growing interest and has inspired the development of benchmarks such as Meta-Dataset. A key challenge in this multi-domain setting is to effectively integrate the feature representations from the diverse set of training domains. Here, we propose a Universal Representation Transformer (URT) layer, that meta-learns to leverage universal features for few-shot classification by dynamically re-weighting and composing the most appropriate domain-specific representations. In experiments, we show that URT sets a new state-of-the-art result on Meta-Dataset. Specifically, it achieves top-performance on the highest number of data sources compared to competing methods. We analyze variants of URT and present a visualization of the attention score heatmaps that sheds light on how the model performs cross-domain generalization. Our code is available at https://github.com/liulu112601/URT.