Disentangling factors of variation in deep representation using adversarial training

Open in new window