Provided are a learning device, a learning method, and a program capable of improving the classification accuracy of images belonging to a target domain. The classification learning unit (74) is based on the comparison result between the output when the feature amount data is input to the classification unit (32) and the teacher data for a plurality of classes associated with the source domain learning image. The learning of the quantity extraction unit (72) and the classification unit (32) is executed. The division unit (50) uses the feature amount data output by the feature amount extraction unit (72) in response to the input of the image as a plurality of partial feature amount data corresponding to the image including one or a plurality of classes of feature maps. Divide into. The domain identification learning unit (76) receives output when the partial feature amount data corresponding to the image is input to the domain identification unit (56), data indicating whether the image belongs to the source domain or the target domain, and data. Based on the comparison result of, the learning of the feature amount extraction unit (72) is executed.
展开▼