# Off-Policy Reinforcement Learning for Efficient and Effective GAN Architecture Search

arXiv: https://arxiv.org/abs/2007.09180 · HTML: https://mlsanity.com/paper/2007.09180/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Jul 2020.

## Built on by (most cited first, 25 of 25)

- [Consistency Models](https://mlsanity.com/paper/2303.01469/) (2023)
- [One-step Diffusion with Distribution Matching Distillation](https://mlsanity.com/paper/2311.18828/) (2023)
- [Extracting Training Data from Diffusion Models](https://mlsanity.com/paper/2301.13188/) (2023)
- [TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up](https://mlsanity.com/paper/2102.07074/) (2021)
- [Neural Architecture Search: Insights from 1000 Papers](https://mlsanity.com/paper/2301.08727/) (2023)
- [Robust Learning Meets Generative Models: Can Proxy Distributions Improve Adversarial Robustness?](https://mlsanity.com/paper/2104.09425/) (2021)
- [Symbolic Regression via Neural-Guided Genetic Programming Population Seeding](https://mlsanity.com/paper/2111.00053/) (2021)
- [Gradient Normalization for Generative Adversarial Networks](https://mlsanity.com/paper/2109.02235/) (2021)
- [PSViT: Better Vision Transformer via Token Pooling and Attention Sharing](https://mlsanity.com/paper/2108.03428/) (2021)
- [Multi-conditioned Graph Diffusion for Neural Architecture Search](https://mlsanity.com/paper/2403.06020/) (2024)
- [Generative Flows with Invertible Attentions](https://mlsanity.com/paper/2106.03959/) (2021)
- [Reinforcement Learning for Generative AI: A Survey](https://mlsanity.com/paper/2308.14328/) (2023)
- [Partition-Guided GANs](https://mlsanity.com/paper/2104.00816/) (2021)
- [Game of GANs: Game-Theoretical Models for Generative Adversarial Networks](https://mlsanity.com/paper/2106.06976/) (2021)
- [SMaRt: Improving GANs with Score Matching Regularity](https://mlsanity.com/paper/2311.18208/) (2023)
- [EAGAN: Efficient Two-stage Evolutionary Architecture Search for GANs](https://mlsanity.com/paper/2111.15097/) (2021)
- [A Prescriptive Dirichlet Power Allocation Policy with Deep Reinforcement Learning](https://mlsanity.com/paper/2201.08445/) (2022)
- [Pareto-aware Neural Architecture Generation for Diverse Computational Budgets](https://mlsanity.com/paper/2210.07634/) (2022)
- [INTAGS: Interactive Agent-Guided Simulation](https://mlsanity.com/paper/2309.01784/) (2023)
- [CycleGANAS: Differentiable Neural Architecture Search for CycleGAN](https://mlsanity.com/paper/2311.07162/) (2023)
- [Dextr: Zero-Shot Neural Architecture Search with Singular Value Decomposition and Extrinsic Curvature](https://mlsanity.com/paper/2508.12977/) (2025)
- [RevoNAD: Reflective Evolutionary Exploration for Neural Architecture Design](https://mlsanity.com/paper/2512.05403/) (2025)
- [Penalty Gradient Normalization for Generative Adversarial Networks](https://mlsanity.com/paper/2306.13576/) (2023)
- [Topology-Aware and Highly Generalizable Deep Reinforcement Learning for Efficient Retrieval in Multi-Deep Storage Systems](https://mlsanity.com/paper/2506.14787/) (2025)
- [Neural Architecture Search for Generative Adversarial Networks: A Comprehensive Review and Critical Analysis](https://mlsanity.com/paper/2606.26169/) (2026)

## Abstract (arXiv)

In this paper, we introduce a new reinforcement learning (RL) based neural architecture search (NAS) methodology for effective and efficient generative adversarial network (GAN) architecture search. The key idea is to formulate the GAN architecture search problem as a Markov decision process (MDP) for smoother architecture sampling, which enables a more effective RL-based search algorithm by targeting the potential global optimal architecture. To improve efficiency, we exploit an off-policy GAN architecture search algorithm that makes efficient use of the samples generated by previous policies. Evaluation on two standard benchmark datasets (i.e., CIFAR-10 and STL-10) demonstrates that the proposed method is able to discover highly competitive architectures for generally better image generation results with a considerably reduced computational burden: 7 GPU hours. Our code is available at https://github.com/Yuantian013/E2GAN.