Parallel/distributed implementation of cellular training for generative\n adversarial neural networks

2020/04/07 by Emiliano Pérez, Perez, Emiliano, Sergio Nesmachnow +7
Computer Science · Physics and Astronomy · #Cellular Automata and Applications #Distributed #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Model Reduction and Neural Networks #Neural and Evolutionary Computing (cs.NE) #Parallel #and Cluster Computing (cs.DC)

paper · pdf · doi:10.48550/arxiv.2004.04633

openalex publication_date 2020/04/07 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28

Abstract

Generative adversarial networks (GANs) are widely used to learn generative\nmodels. GANs consist of two networks, a generator and a discriminator, that\napply adversarial learning to optimize their parameters. This article presents\na parallel/distributed implementation of a cellular competitive coevolutionary\nmethod to train two populations of GANs. A distributed memory parallel\nimplementation is proposed for execution in high performance/supercomputing\ncenters. Efficient results are reported on addressing the generation of\nhandwritten digits (MNIST dataset samples). Moreover, the proposed\nimplementation is able to reduce the training times and scale properly when\nconsidering different grid sizes for training.\n

Related

• A New Distributed Method for Training Generative Adversarial Networks
• A Decentralized Parallel Algorithm for Training Generative Adversarial Nets
• Lipizzaner: A System That Scales Robust Generative Adversarial Network Training
• Generative Adversarial Parallelization
• A Distributed Training Algorithm of Generative Adversarial Networks with Quantized Gradients