Generative adversarial interpolative autoencoding: adversarial training on latent space interpolations encourage convex latent distributions

Abstract

We present a neural network architecture based upon the Autoencoder (AE) and Generative Adversarial Network (GAN) that promotes a convex latent distribution by training adversarially on latent space interpolations. By using an AE as both the generator and discriminator of a GAN, we pass a pixel-wise error function across the discriminator, yielding an AE which produces non-blurry samples that match both high- and low-level features of the original images. Interpolations between images in this space remain within the latent-space distribution of real images as trained by the discriminator, and therfore preserve realistic resemblances to the network inputs. Code available at https://github.com/timsainb/GAIA

References in corpus (15)

• Scikit-learn: Machine Learning in Python
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• WaveNet: A Generative Model for Raw Audio
• Pixel Recurrent Neural Networks
• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• NICE: Non-linear Independent Components Estimation
• Density estimation using Real NVP
• BEGAN: Boundary Equilibrium Generative Adversarial Networks
• Adversarially Learned Inference
• Autoencoding beyond pixels using a learned similarity metric
• Invertible Conditional GANs for image editing
• Neural Audio Synthesis of Musical Notes with WaveNet Autoencoders
• Multimodal Unsupervised Image-to-Image Translation
• Deep Identity-aware Transfer of Facial Attributes
• One-Sided Unsupervised Domain Mapping

Cited by in corpus (15)

• A survey on GANs for computer vision: Recent research, analysis and taxonomy
• Anomaly Detection with Adversarial Dual Autoencoders
• Deep learning for nano-photonic materials -- The solution to everything!?
• Unsupervised Anomaly Detection with Adversarial Mirrored AutoEncoders
• Autoencoder Image Interpolation by Shaping the Latent Space
• Conditional Adversarial Generative Flow for Controllable Image Synthesis
• CMTS: Conditional Multiple Trajectory Synthesizer for Generating Safety-critical Driving Scenarios
• Connect the Dots: In Situ 4D Seismic Monitoring of CO2 Storage with Spatio-temporal CNNs
• Learning a low dimensional manifold of real cancer tissue with PathologyGAN
• Classify and Generate: Using Classification Latent Space Representations for Image Generations
• Uniform Interpolation Constrained Geodesic Learning on Data Manifold
• Augmentation-Interpolative AutoEncoders for Unsupervised Few-Shot Image Generation
• Covering up bias in CelebA-like datasets with Markov blankets: A post-hoc cure for attribute prior avoidance
• Encoded Prior Sliced Wasserstein AutoEncoder for learning latent manifold representations
• Deep Unsupervised Common Representation Learning for LiDAR and Camera Data using Double Siamese Networks