MLLGJan 9, 2017

AdaGAN: Boosting Generative Models

arXiv:1701.02386v2230 citations
Originality Incremental advance
AI Analysis

This addresses the training instability and mode collapse issue in GANs for generative modeling, representing an incremental improvement with a novel hybrid approach.

The paper tackles the problem of missing modes in Generative Adversarial Networks (GANs) by proposing AdaGAN, an iterative boosting-inspired method that adds components to a mixture model using reweighted samples, which experimentally addresses mode collapse and theoretically ensures convergence.

Generative Adversarial Networks (GAN) (Goodfellow et al., 2014) are an effective method for training generative models of complex data such as natural images. However, they are notoriously hard to train and can suffer from the problem of missing modes where the model is not able to produce examples in certain regions of the space. We propose an iterative procedure, called AdaGAN, where at every step we add a new component into a mixture model by running a GAN algorithm on a reweighted sample. This is inspired by boosting algorithms, where many potentially weak individual predictors are greedily aggregated to form a strong composite predictor. We prove that such an incremental procedure leads to convergence to the true distribution in a finite number of steps if each step is optimal, and convergence at an exponential rate otherwise. We also illustrate experimentally that this procedure addresses the problem of missing modes.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes