Disentangling factors of variation in deep representations using adversarial training

Open in new window