---
title: Progressive Growing of GANs
url: https://www.emergentmind.com/topics/progressive-growing-of-gans-p-gan
type: topic
---

# Progressive Growing of GANs

Progressive Growing of Generative Adversarial Networks (P-GAN) is a training methodology in which a GAN's generator and discriminator are both repeatedly extended by new layers, allowing the networks to synthesize increasingly fine-scale detail as resolution is gradually increased. This approach, popularized by Karras et al., has led to marked advances in high-resolution image synthesis and has been adapted into medical imaging, video synthesis, physical layout generation, and other domains. By decomposing the generative task into a sequence of sub-tasks that tackle ever-greater spatial (and sometimes temporal) granularity, P-GAN improves stability, realism, and variation while remaining tractable for modern hardware.

## 1. Foundational Principles and Progressive Growing Strategy

At the core of P-GAN is the progressive layering of both generator ($G$) and discriminator ($D$), who begin operating at a minimal spatial scale (typically $4 \times 4$ or $8 \times 8$). After initial convergence at this low resolution, a new block is appended to both $G$ and $D$, doubling the working resolution (e.g., $8 \to 16$, $16 \to 32$, etc.) [1710.10196], [1805.03144], [1911.13029].

During each transition, the networks employ a linear fade-in mechanism controlled by a scalar $\alpha \in [0,1]$, which blends activations from the old layers (upsampled path) and new block (convolution path) according to
\[
\text{output}_{2R} = \alpha \cdot \text{new\_path}_{2R} + (1 - \alpha) \cdot \text{old\_path}_{2R}
\]
with $\alpha$ increasing over a prescribed number of images or iterations [1710.10196], [1911.13029]. Once $\alpha=1$, training proceeds on the fully grown network for "stabilization" before the next resolution increase.

This smooth transition is crucial to retain previously learned coarse features and to prevent destabilization as resolution increases.

## 2. Architectural Components and Training Workflow

The generator structure at each stage consists of an upsampling operation, followed by pairs of $3\times3$ convolutional layers, pixelwise feature normalization, and nonlinear activation (usually LeakyReLU). Correspondingly, the discriminator mirrors this structure with $3\times3$ convolutions followed by average pooling (downsampling), a minibatch standard deviation layer for mode collapse mitigation, and final fully connected layers for scalar output [1710.10196], [1805.03144], [1912.05357].

A weight scaling ("equalized learning rate") mechanism is employed per layer, standardizing the dynamic range across layers. The feature map scheduling typically follows:

| Resolution | Generator: Feature Maps | Discriminator: Feature Maps |
|------------|------------------------|----------------------------|
| 4x4        | 512                    | 512                        |
| 8x8        | 512                    | 512                        |
| 16x16      | 512                    | 512                        |
| 32x32      | 512                    | 512                        |
| 64x64      | 256                    | 256                        |
| 128x128    | 128                    | 128                        |
| 256x256    | 64                     | 64                         |
| 512x512    | 32                     | 32                         |
| 1024x1024  | 16                     | 16                         |

Pixelwise feature normalization takes place after each activation in the generator:

\[
b_{x,y,i} = \frac{a_{x,y,i}}{\sqrt{\frac{1}{N} \sum_{j=1}^N a_{x,y,j}^2 + \epsilon}}
\]
where $\epsilon=10^{-8}$ and $N$ is the number of channels.

For variants focused on computational efficiency, all standard convolutions may be replaced by depthwise-separable convolution blocks, resulting in markedly reduced multiply–accumulate counts and approximately a $2\times$ speed-up per training epoch at $64\times64$ resolution [2202.12337].

## 3. Loss Functions, Training Schedules, and Stability Mechanisms

P-GAN almost universally employs the Wasserstein GAN loss with gradient penalty (WGAN-GP):

\[
L_D = \mathbb{E}_{\tilde{x}\sim\mathbb{P}_g}[D(\tilde{x})] - \mathbb{E}_{x\sim\mathbb{P}_r}[D(x)] + \lambda\,\mathbb{E}_{\hat{x}\sim\mathbb{P}_{\hat{x}}}\left(\|\nabla_{\hat{x}}D(\hat{x})\|_{2}-1\right)^2
\]
\[
L_G = -\mathbb{E}_{z\sim p(z)}\left[D\left(G(z)\right)\right]
\]
where $\lambda$ is a penalty coefficient, $p(z)$ is the latent prior, and $\mathbb{P}_r$, $\mathbb{P}_g$ denote real and generated distributions [1710.10196], [1911.13029], [2202.12337]. The optimization utilizes Adam, commonly with $\beta_1=0$, $\beta_2=0.99$, and dynamically adjusted batch sizes to fit available GPU memory.

Stability-enabling tricks include pixelwise feature vector normalization in $G$ and per-feature minibatch standard deviation aggregation in $D$, which mitigates unhealthy generator/discriminator competition and mode collapse [1710.10196], [1805.03144].

In settings where progression is truncated at intermediate resolutions, outputs are post-processed by a super-resolution GAN (SRGAN), which is independently trained to upsample images. The composite SRGAN loss is

\[
L_{SR} = L_{content} + \beta \, L_{adv,SR}
\]
with perceptual content loss based on VGG features and adversarial loss with a small $\beta$ weight [2202.12337].

## 4. Domain Adaptations and Extensions

The progressive growing methodology has been extended well beyond natural image synthesis. In medical imaging, segmentation maps are incorporated as extra input or output channels in $G$ and $D$, enabling the networks to synthesize anatomical and pathological detail at native resolution [1805.03144], [2004.00226]. In video synthesis, 3D convolutional blocks are progressively grown in both spatial and temporal directions, with dedicated fade-in schedules for each dimension [1810.02419], [1810.01325].

In physical layout optimization (metasurfaces), conditional inputs (e.g. wavelength, deflection angle) are embedded and concatenated to the noise vector $z$ before feeding to $G$, enabling parametric image generation. Progressive training-set refinement is used (see "UpdateTrainingSet" pseudocode in [1911.13029]), yielding significant computational cost reductions over conventional topology optimization.

For three-dimensional neuroimaging synthesis, all operations are replaced by 3D variants, channel counts are reduced to fit memory constraints, and data volumes are cropped and upsampled appropriately [1912.05357].

## 5. Evaluation Metrics, Experimental Results, and Benchmarks

Progressive growing improves both image fidelity and diversity. Key evaluation metrics include Sliced Wasserstein Distance (SWD), Fréchet Inception Distance (FID), Mean Structural SIMilarity index (MS-SSIM), and application-specific measures (AUC on vessels, dice score on segmentations, deflection efficiency for metasurfaces):

| Method                  | SWD (patch) | FID    | MS-SSIM   | Inception Score (IS) |
|-------------------------|-------------|---------|-----------|---------------------|
| P-GAN (full model)      | 2.96e-3     | 8.34    | 0.2828    | 8.80 (CIFAR-10)     |
| Sketch-guided PGSGAN    | --          | 54.94   | 0.4895    | --                  |
| P-GAN + SRGAN (CelebA)  | 381.86      | --      | 0.1698    | 2.138 ± 0.130       |

User studies and downstream segmentation tasks confirm that realism and utility are substantially increased relative to both non-progressive GANs and conventional adversarial architectures [2004.00226], [1805.03144].

Training costs depend heavily on resolution and architecture. For $64\times64$ images, DS Convs halve the per-epoch time; for $512\times512$ synthesis, staging progressive growth avoids days of training times at the highest resolutions [2202.12337], [1805.03144].

## 6. Practical Considerations and Implementation Guidelines

A range of practical issues must be addressed for efficient and stable P-GAN deployment:
- GPU memory consumption increases rapidly with resolution; reductions in batch size and freezing of early layers may be necessary [1805.03144], [1912.05357].
- Fade-in schedules should be sufficiently long (e.g., 20,000 batches or more per phase) to avoid destabilization [1805.03144].
- Equalized learning rate and pixelwise normalization are preferred over batchnorm [1710.10196].
- Poor segmentation or conditional labels will degrade output quality in conditional and segmentation-aware variants [1805.03144].
- Transfer learning is facilitated by retaining low-resolution weights and restarting progression at an appropriate stage for a new dataset [1805.03144].
- Task-specific output and conditioning strategies can be implemented provided the progressive schedule is respected [1911.13029].

## 7. Extensions, Limitations, and Future Directions

Current adaptations of progressive growing cover: image synthesis up to $1024^2$ resolution [1710.10196], medical image domains with auxiliary segmentation [1805.03144], ultrasound and other modalities with sketch guidance [2004.00226], 3D neuroimaging synthesis up to $64^3$ voxels [1912.05357], and video sequences with spatio-temporal resolution rising jointly [1810.02419], [1810.01325].

Reported limitations include hardware-imposed resolution ceilings (e.g., $64^3$ for 3D volumes), lack of quantitative metrics in certain domains (neuroimaging, video), and eventual saturation in image realism at extreme scales. Future work encompasses extending P-GAN to conditional domains, enhancing training-set refinement, and pushing resolution and temporal scales further with distributed computation and improved memory management [1912.05357], [1810.02419], [1911.13029].

Through a curriculum-based regime of incremental resolution growth, progressive GANs have provided a stable and high-fidelity foundation for generative modeling in a wide array of high-dimensional domains.

Source: https://www.emergentmind.com/topics/progressive-growing-of-gans-p-gan