# Signal Propagation in a Gradient-Based and Evolutionary Learning System

arXiv: https://arxiv.org/abs/2102.08929 · HTML: https://mlsanity.com/paper/2102.08929/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 33 references that are in the map.
- First posted on arXiv 10 Feb 2021.

## Builds on (most cited first, 23 of 23)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Progressive Growing of GANs for Improved Quality, Stability, and Variation](https://mlsanity.com/paper/1710.10196/) (2017)
- [Improved Techniques for Training GANs](https://mlsanity.com/paper/1606.03498/) (2016)
- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks](https://mlsanity.com/paper/1511.06434/) (2015)
- [Least Squares Generative Adversarial Networks](https://mlsanity.com/paper/1611.04076/) (2016)
- [Wasserstein GAN](https://mlsanity.com/paper/1701.07875/) (2017)
- [Towards Principled Methods for Training Generative Adversarial Networks](https://mlsanity.com/paper/1701.04862/) (2017)
- [The relativistic discriminator: a key element missing from standard GAN](https://mlsanity.com/paper/1807.00734/) (2018)
- [Generalization and Equilibrium in Generative Adversarial Nets (GANs)](https://mlsanity.com/paper/1703.00573/) (2017)
- [Energy-based Generative Adversarial Network](https://mlsanity.com/paper/1609.03126/) (2016)
- [COVID-19 Image Data Collection](https://mlsanity.com/paper/2003.11597/) (2020)
- [Dual Discriminator Generative Adversarial Nets](https://mlsanity.com/paper/1709.03831/) (2017)
- [Evolutionary Generative Adversarial Networks](https://mlsanity.com/paper/1803.00657/) (2018)
- [Spatial Evolutionary Generative Adversarial Networks](https://mlsanity.com/paper/1905.12702/) (2019)
- [Towards Distributed Coevolutionary GANs](https://mlsanity.com/paper/1807.08194/) (2018)
- [Coevolution of Generative Adversarial Networks](https://mlsanity.com/paper/1912.06172/) (2019)
- [Data Dieting in GAN Training](https://mlsanity.com/paper/2004.04642/) (2020)
- [Lipizzaner: A System That Scales Robust Generative Adversarial Network Training](https://mlsanity.com/paper/1811.12843/) (2018)
- [Distributional Adversarial Networks](https://mlsanity.com/paper/1706.09549/) (2017)
- [Re-purposing Heterogeneous Generative Ensembles with Evolutionary Computation](https://mlsanity.com/paper/2003.13532/) (2020)
- [Analyzing the Components of Distributed Coevolutionary GAN Training](https://mlsanity.com/paper/2008.01124/) (2020)

## Built on by (most cited first, 3 of 3)

- [Combating COVID-19 using Generative Adversarial Networks and Artificial Intelligence for Medical Images: A Scoping Review](https://mlsanity.com/paper/2205.07236/) (2022)
- [Leveraging GANs for data scarcity of COVID-19: Beyond the hype](https://mlsanity.com/paper/2304.03536/) (2023)
- [Population-Based Multi-Objective Training of Discriminators for Semi-Supervised GANs](https://mlsanity.com/paper/2607.01907/) (2026)

## Abstract (arXiv)

Generative adversarial networks (GANs) exhibit training pathologies that can lead to convergence-related degenerative behaviors, whereas spatially-distributed, coevolutionary algorithms (CEAs) for GAN training, e.g. Lipizzaner, are empirically robust to them. The robustness arises from diversity that occurs by training populations of generators and discriminators in each cell of a toroidal grid. Communication, where signals in the form of parameters of the best GAN in a cell propagate in four directions: North, South, West, and East, also plays a role, by communicating adaptations that are both new and fit. We propose Lipi-Ring, a distributed CEA like Lipizzaner, except that it uses a different spatial topology, i.e. a ring. Our central question is whether the different directionality of signal propagation (effectively migration to one or more neighbors on each side of a cell) meets or exceeds the performance quality and training efficiency of Lipizzaner Experimental analysis on different datasets (i.e, MNIST, CelebA, and COVID-19 chest X-ray images) shows that there are no significant differences between the performances of the trained generative models by both methods. However, Lipi-Ring significantly reduces the computational time (14.2%. . . 41.2%). Thus, Lipi-Ring offers an alternative to Lipizzaner when the computational cost of training matters.