Fetching the paper…

Disentangling factors of variation in deep representations using adversarial training · Around