Distributional Adversarial Networks
arXiv:1706.09549
Abstract
We propose a framework for adversarial training that relies on a sample rather than a single sample point as the fundamental unit of discrimination. Inspired by discrepancy measures and two-sample tests between probability distributions, we propose two such distributional adversaries that operate and predict on samples, and show how they can be easily implemented on top of existing models. Various experimental results show that generators trained with our distributional adversaries are much more stable and are remarkably less prone to mode collapse than traditional models trained with pointwise prediction discriminators. The application of our framework to domain adaptation also results in considerable improvement over recent state-of-the-art.
References in corpus (10)
- Adam: A Method for Stochastic Optimization
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
- Energy-based Generative Adversarial Network
- Generating Videos with Scene Dynamics
- Adversarial Discriminative Domain Adaptation
- Unrolled Generative Adversarial Networks
- Mode Regularized Generative Adversarial Networks
- Central Moment Discrepancy (CMD) for Domain-Invariant Representation Learning
- Quality Aware Network for Set to Set Recognition
Cited by in corpus (8)
- Demystifying MMD GANs
- Spatial Evolutionary Generative Adversarial Networks
- Lipizzaner: A System That Scales Robust Generative Adversarial Network Training
- Large Scale Many-Objective Optimization Driven by Distributional Adversarial Networks
- TripletGAN: Training Generative Model with Triplet Loss
- Inception Score, Label Smoothing, Gradient Vanishing and -log(D(x)) Alternative
- Understanding the Effectiveness of Lipschitz-Continuity in Generative Adversarial Nets
- Dist-GAN: An Improved GAN using Distance Constraints