REVIEW 2 cited by
Combating Mode Collapse in GAN training: An Empirical Analysis using Hessian Eigenvalues
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
abstract
Generative adversarial networks (GANs) provide state-of-the-art results in image generation. However, despite being so powerful, they still remain very challenging to train. This is in particular caused by their highly non-convex optimization space leading to a number of instabilities. Among them, mode collapse stands out as one of the most daunting ones. This undesirable event occurs when the model can only fit a few modes of the data distribution, while ignoring the majority of them. In this work, we combat mode collapse using second-order gradient information. To do so, we analyse the loss surface through its Hessian eigenvalues, and show that mode collapse is related to the convergence towards sharp minima. In particular, we observe how the eigenvalues of the $G$ are directly correlated with the occurrence of mode collapse. Finally, motivated by these findings, we design a new optimization algorithm called nudged-Adam (NuGAN) that uses spectral information to overcome mode collapse, leading to empirically more stable convergence properties.
Forward citations
Cited by 2 Pith papers
-
ROMS-IMLE: A Minimalist Approach to Competitive Single-Step Generative Modelling
A single-step IMLE generator with per-stage supervision and a robust loss reports FID 2.56 on ImageNet-256 by filtering ~5% of samples at test time.
-
Generative Adversarial Networks Bridging Art and Machine Intelligence
This paper is a textbook-style review of generative adversarial networks, covering theory, classic variants, training methods, and applications; no new architecture, theorem, or experimental result is introduced.
Discussion (0). Continue with ORCID to comment.