Catastrophic forgetting and mode collapse in GANs

By Hoang Thanh-Tung, Truyen Tran. 2020

283 citations recorded by the source corpus.

In this paper, we show that Generative Adversarial Networks (GANs) suffer from catastrophic forgetting even when they are trained to approximate a single target distribution. We show that GAN training is a continual learning problem in which the sequence of changing model distributions is the sequence of tasks to the discriminator. The level of mismatch between tasks in the sequence determines the level of forgetting. Catastrophic forgetting is interrelated to mode collapse and can make the training of GANs non-convergent. We investigate the landscape of the discriminator’s output in different variants of GANs and find that when a GAN converges to a good equilibrium, real training datapoints are wide local maxima of the discriminator. We empirically show the relationship between the sharpness of local maxima and mode collapse and generalization in GANs. We show how catastrophic forgetting prevents the discriminator from making real datapoints local maxima, and thus causes non-convergence. Finally, we study methods for preventing catastrophic forgetting in GANs.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Generative Models, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Generative Models

In Safety, Robustness & Interpretability

Citations in the corpus

A Comprehensive Survey of Continual Learning: Theory, Method and Application (2023, background): “Similarly, resolving within-task catastrophic forgetting can facilitate reinforcement learning [299], [300] and stabilize the training of GANs [301], [302].”
Task Agnostic Continual Learning via Meta Learning (2019, methodology): “Indeed, it has been shown empirically by Thanh-Tung & Tran (2019); Liang et al. (2019) that GANs suffer from catastrophic forgetting, and they adapted CL methods such as Elastic Weight Consolidation (EWC) (Kirkpatrick et al., 2017) and Synaptic Intelligence (SI) (Zenke et al., 2017) to alleviate…”
Denoising diffusion probabilistic models for 3D medical image generation (2023, unknown): “GANs often do not converge during training or suffer from mode collapse 6 —a phenomenon we did not observe for our latent diffusion model.”
On Data Augmentation for GAN Training (2020, background): “This constraint mitigates gradient vanishing [24] and catastrophic forgetting [31].”
StyleAlign: Analysis and Applications of Aligned StyleGAN Models (2021, background): “CF has been also studied in the context of GANs (Liang et al., 2018; Li et al., 2020; Thanh-Tung & Tran, 2020).”
Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Models (2021, background): “The discriminator can also jump between modes resulting in catastrophic forgetting, where previously learned knowledge is forgotten when learning something new [211].”
ECGTwin: Personalized ECG Generation Using Controllable Diffusion Model (2025, unknown): “…the abundance of conditions offers the potential for fine-grained control, it also poses a significant challenge: if not properly integrated, these conditions may confuse the generative model, leading to degraded performance or even model collapse (Fetaya et al. 2020; Thanh-Tung and Tran 2020).”