Adversarially Learned Inference

Vincent Dumoulin, Ishmael Belghazi, Ben Poole, Alex Lamb, Martin Arjovsky, Olivier Mastropietro, Aaron Courville

Nov 04, 2016 (modified: Feb 21, 2017) ICLR 2017 conference submission readers: everyone
  • Abstract: We introduce the adversarially learned inference (ALI) model, which jointly learns a generation network and an inference network using an adversarial process. The generation network maps samples from stochastic latent variables to the data space while the inference network maps training examples in data space to the space of latent variables. An adversarial game is cast between these two networks and a discriminative network that is trained to distinguish between joint latent/data-space samples from the generative network and joint samples from the inference network. We illustrate the ability of the model to learn mutually coherent inference and generation networks through the inspections of model samples and reconstructions and confirm the usefulness of the learned representations by obtaining a performance competitive with other recent approaches on the semi-supervised SVHN task.
  • TL;DR: We present and adverserially trained generative model with an inference network. Samples quality is high. Competitive semi-supervised results are achieved.
  • Conflicts:
  • Keywords: Computer vision, Deep learning, Unsupervised Learning, Semi-Supervised Learning