English Dictionary

zero-shot learning

名詞

不可算名詞

(computing) A problem setup in machine learning, where, at test time, a learner observes samples from classes that were not observed during training and attempts to predict the class that they belong to, typically based on auxiliary information.

日本語の意味

（機械学習における）学習時に観測されなかったクラスのデータがテスト時に現れ、それに基づき補助情報などを利用して新たなクラスを予測する問題設定または手法。

In many recent benchmarks, zero-shot learning enables models to classify images of categories that were never seen during training by leveraging attribute descriptions.

多くの最新ベンチマークでは、ゼロショット学習によりモデルが訓練時に一度も観測されなかったカテゴリの画像を属性記述を活用して分類できるようになります。

意味（1）

(computing) A problem setup in machine learning, where, at test time, a learner observes samples from classes that were not observed during training and attempts to predict the class that they belong to, typically based on auxiliary information.