论文标题
通过未标记的示例增强几个图像分类
Enhancing Few-Shot Image Classification with Unlabelled Examples
论文作者
论文摘要
我们开发了一种跨传输元学习方法，该方法使用未标记的实例来改善几乎没有图像的分类性能。我们的方法结合了基于Mahalanobis-Distance的正规化软K-均值聚类程序和改良的最先进的神经适应性特征提取器，以使用未标记的数据来提高测试时间分类精度。我们评估了我们的跨导次学习任务的方法，其中的目标是在给定一组支持（培训）示例的情况下共同预测查询（测试）示例的标签。我们在Meta-Dataset，Mini-Imagenet和分层Imagenet基准上实现了最先进的性能。所有训练有素的模型和代码均在github.com/plai-group/simple-cnaps上公开提供。
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.
