Papers › Enhancing GAN Performance through Neural Architecture Search and Tensor Decomposition
Enhancing GAN Performance through Neural Architecture Search and Tensor Decomposition
Prasanna Reddy Pulakurthi, Mahsa Mozaffari, Sohail A. Dianat, Majid Rabbani, Jamison Heard, Raghuveer Rao
Generative Adversarial Networks (GANs) have emerged as a powerful tool for generating high-fidelity content. This paper presents a new training procedure that leverages Neural Architecture Search (NAS) to discover the optimal architecture for image generation while employing the Maximum Mean Discrepancy (MMD) repulsive loss for adversarial training. Moreover, the generator network is compressed using tensor decomposition to reduce its computational footprint and inference time while preserving its generative performance. Experimental results show improvements of 34% and 28% in the FID score on the CIFAR-10 and STL-10 datasets, respectively, with corresponding footprint reductions of 14× and 31× compared to the best FID score method reported in the literature. The implementation code is available at: https://github.com/PrasannaPulakurthi/MMD-AdversarialNAS.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Image Generation | STL-10 | MMD-AdversarialNAS | FID | 12.91 | #6 of 31 | Archive leaderboard | report |
| Image Generation | STL-10 | MMD-AdversarialNAS | Inception score | 11.6 | #6 of 31 | Archive leaderboard | report |
| Image Generation | STL-10 | MMD-AdversarialNAS | Model Size (MB) | 19.47 | #6 of 31 | Archive leaderboard | report |
| Image Generation | STL-10 | MMD-AdversarialNAS (Compressed Large) | FID | 13.06 | #8 of 31 | Archive leaderboard | report |
| Image Generation | STL-10 | MMD-AdversarialNAS (Compressed Large) | Inception score | 11.28 | #8 of 31 | Archive leaderboard | report |
| Image Generation | STL-10 | MMD-AdversarialNAS (Compressed Large) | Model Size (MB) | 5.35 | #8 of 31 | Archive leaderboard | report |
| Image Generation | STL-10 | MMD-AdversarialNAS (Compressed Small) | FID | 14.84 | #11 of 31 | Archive leaderboard | report |
| Image Generation | STL-10 | MMD-AdversarialNAS (Compressed Small) | Inception score | 11.66 | #11 of 31 | Archive leaderboard | report |
| Image Generation | STL-10 | MMD-AdversarialNAS (Compressed Small) | Model Size (MB) | 1.71 | #11 of 31 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections