We develop a transductive meta-learning method that uses unlabelled instances
to improve few-shot image classification performance. Our approach combines a
regularized Mahalanobis-distance-based soft k-means clustering procedure with a
modified state of the art neural adaptive feature extractor to achieve improved
test-time classification accuracy using unlabelled data. We evaluate our method
on transductive few-shot learning tasks, in which the goal is to jointly
predict labels for query (test) examples given a set of support (training)
examples. We achieve state of the art performance on the Meta-Dataset,
mini-ImageNet and tiered-ImageNet benchmarks. All trained models and code have
been made publicly available at github.com/plai-group/simple-cnaps.

我们提出了一种用于少样本图像分类的传导元学习方法，结合了经过正则化的马氏距离软 k-means 聚类过程和修改后的最先进的神经自适应特征提取器来使用未标记数据提高测试时间分类准确性，我们在 Meta-Dataset、mini-ImageNet 和 tiered-ImageNet 基准测试上取得了最先进的性能。