Also known as: Zero shot Learning · Zero Shot Learning
Zero-shot learning — это способность машинного обучения, при которой модель выполняет задачу или распознает категорию без task-specific training examples. Она опирается на general knowledge, learned representations или relationships from prior training, чтобы generalize to unseen tasks or concepts. В некоторых контекстах термин используется более узко для extracting или classifying information without annotated examples.
A type of machine learning training where the model infers a prediction for a task that it was not specifically already trained on. In other words, the model is given zero task-specific training examples but asked to do inference for that task.