Utilizing Information from Task-Independent Aspects via GAN-Assisted Knowledge Transfer

Publication Type:
Conference Proceeding
Proceedings of the International Joint Conference on Neural Networks, 2018, 2018-July
Issue Date:
Filename Description Size
08489047.pdfPublished version1.9 MB
Adobe PDF
Full metadata record
© 2018 IEEE. Observed data often have multiple labels with respect to different aspects. For example, a picture can have one label specifying the contents in terms of the object category such as aeroplane, building, cat, etc. And in the meanwhile have another label describing the image style such as photo-realistic or artistic. The central idea of this work is that any annotation of the data contains precious knowledge and is not to be foregone: An analytic task focusing on one aspect of the data can benefit from the knowledge transferred from the other aspects. We propose a passive knowledge transfer scheme for deep neural network training based on the generative adversarial nets (GANs). The adversarial training scheme encourages the nets to encode data into representations that are both discriminative for the target aspect and invariant with respect to the irrelevant aspects. We show that the scheme mixes the conditional distributions of the encoded data on the irrelevant aspects, by the theory on the link between the GAN framework and the Wasserstein metric in distribution spaces. Moreover, we empirically verified the method by i) classifying images despite influence by geometric transform and ii) recognizing the movements (geometric transform) regardless the image contents.
Please use this identifier to cite or link to this item: