AI Paper Atlas

Research map / Image generation / GANs

Learning Energy-Based Models with Adversarial Training

Xuwang Yin, Shiying Li, Gustavo K. Rohde

ECCV 2022 · Image generation › GANs · approach: Generative adversarial network theory · 6 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use a more diverse p0 dataset (ablation)binary AT generative model · CIFAR-10FIDImproved
increase the number of PGD steps K during training (ablation)binary AT generative model · CIFAR-10FIDImproved
progressively increase K from 0 to 25 instead of fixed K=25 (ablation)binary AT generative model · CIFAR-10training convergence speedImproved
pretrain the discriminator on ImageNet classification (ablation)binary AT generative model · CelebA-HQ 256FIDImproved
add R1 regularization (30) when using a pretrained model (ablation)binary AT generative model · CelebA-HQ 256FIDImproved
add R1 regularization when training from scratch (ablation)binary AT generative model · CelebA-HQ 256FIDNo clear effect
vary number of sampling steps with fixed total perturbation (64x0.1, 32x0.2, 16x0.4) (ablation)binary AT generative model · CIFAR-10FID: 13.07, 13.21, 13.49No clear effect
vary number of sampling steps with fixed total perturbation (40x4.0, 20x8.0, 10x16.0) (ablation)binary AT generative model · CelebA-HQ 256FID: 19.19, 18.97, 19.19No clear effect
apply Gaussian smoothing (sigma=10) to source p0 images (ablation)binary AT generative model · CelebA-HQ 256visual artifacts and FIDMixed
use the EBM likelihood gradient with the PGD sampler instead of the binary AT objective (ablation)binary AT generative model · training stabilitytraining stabilityHurt
binary AT versus explicit EBMs and AT-based modelsbinary AT generative model · CIFAR-10FID: 38.2 -> 13.21Improved
binary AT versus VAEBM on CelebA-HQ 256binary AT generative model · CelebA-HQ 256FID: 20.38 -> 17.31Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy