Alternating Back-Propagation for Generator Network

Abstract

This paper proposes an alternating back-propagation algorithm for learning the generator network model. The model is a non-linear generalization of factor analysis. In this model, the mapping from the continuous latent factors to the observed signal is parametrized by a convolutional neural network. The alternating back-propagation algorithm iterates the following two steps: (1) Inferential back-propagation, which infers the latent factors by Langevin dynamics or gradient descent. (2) Learning back-propagation, which updates the parameters given the inferred latent factors by gradient descent. The gradient computations in both steps are powered by back-propagation, and they share most of their code in common. We show that the alternating back-propagation algorithm can learn realistic generator models of natural images, video sequences, and sounds. Moreover, it can also be used to learn from incomplete or indirect training data.

Cited by in corpus (21)

• Your GAN is Secretly an Energy-based Model and You Should use Discriminator Driven Latent Sampling
• Learned imaging with constraints and uncertainty quantification
• Dense Uncertainty Estimation
• Uncertainty Inspired RGB-D Saliency Detection
• Learning Noise-Aware Encoder-Decoder from Noisy Labels by Alternating Back-Propagation for Saliency Detection
• Learning Deep Generative Models with Annealed Importance Sampling
• Ergodic Inference: Accelerate Convergence by Optimisation
• Cooperative Training of Fast Thinking Initializer and Slow Thinking Solver for Conditional Learning
• Learning Latent Space Energy-Based Prior Model
• Conditional Generative ConvNets for Exemplar-based Texture Synthesis
• Learning Cycle-Consistent Cooperative Networks via Alternating MCMC Teaching for Unsupervised Cross-Domain Translation
• Gradient Origin Networks
• Learning Energy-based Spatial-Temporal Generative ConvNets for Dynamic Patterns
• Deep Unsupervised Clustering with Clustered Generator Model
• A manifold learning perspective on representation learning: Learning decoder and representations without an encoder
• Generative Text Modeling through Short Run Inference
• Inducing Hierarchical Compositional Model by Sparsifying Generator Network
• Mitigating Generation Shifts for Generalized Zero-Shot Learning
• A General Divergence Modeling Strategy for Salient Object Detection
• Representation Learning: A Statistical Perspective
• Dense Uncertainty Estimation via an Ensemble-based Conditional Latent Variable Model