Sparsity through evolutionary pruning prevents neuronal networks from overfitting

Richard C. Gerum, André Erpenbeck, Patrick Krauss, Achim Schilling*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Modern Machine learning techniques take advantage of the exponentially rising calculation power in new generation processor units. Thus, the number of parameters which are trained to solve complex tasks was highly increased over the last decades. However, still the networks fail – in contrast to our brain – to develop general intelligence in the sense of being able to solve several complex tasks with only one network architecture. This could be the case because the brain is not a randomly initialized neural network, which has to be trained from scratch by simply investing a lot of calculation power, but has from birth some fixed hierarchical structure. To make progress in decoding the structural basis of biological neural networks we here chose a bottom-up approach, where we evolutionarily trained small neural networks in performing a maze task. This simple maze task requires dynamic decision making with delayed rewards. We were able to show that during the evolutionary optimization random severance of connections leads to better generalization performance of the networks compared to fully connected networks. We conclude that sparsity is a central property of neural networks and should be considered for modern Machine learning approaches.

Original languageEnglish
Pages (from-to)305-312
Number of pages8
JournalNeural Networks
Volume128
DOIs
StatePublished - Aug 2020

Keywords

  • Artificial neural networks
  • Biological plausibility
  • Evolution
  • Evolutionary algorithm
  • Maze task
  • Overfitting

Fingerprint

Dive into the research topics of 'Sparsity through evolutionary pruning prevents neuronal networks from overfitting'. Together they form a unique fingerprint.

Cite this