Back to Search
Start Over
Training spiking neural networks with a multi-agent evolutionary robotics framework
- Source :
- GECCO
- Publication Year :
- 2021
- Publisher :
- ACM, 2021.
-
Abstract
- We demonstrate the training of Spiking Neural Networks (SNN) in a novel multi-agent Evolutionary Robotics (ER) framework inspired by competitive evolutionary environments in nature. The topology of a SNN along with morphological parameters of the bot it controls in the ER environment is together treated as a phenotype. Rules of the framework select certain bots and their SNNs for reproduction and others for elimination based on their efficacy in capturing food in a competitive environment. While the bots and their SNNs are not explicitly trained to survive or reproduce using loss functions, these drives emerge implicitly as they evolve to hunt food and survive. Their efficiency in capturing food exhibits the evolutionary signature of punctuated equilibrium. We use this signature to compare the performances of two evolutionary inheritance algorithms on the phenotypes, Mutation and Crossover with Mutation, using ensembles of 100 experiments for each algorithm. We find that Crossover with Mutation promotes 40% faster learning in the SNN than mere Mutation with a statistically significant margin.
- Subjects :
- Spiking neural network
Punctuated equilibrium
business.industry
Computer science
media_common.quotation_subject
Crossover
Evolutionary robotics
Inheritance (genetic algorithm)
0102 computer and information sciences
02 engineering and technology
01 natural sciences
010201 computation theory & mathematics
Margin (machine learning)
Mutation (genetic algorithm)
0202 electrical engineering, electronic engineering, information engineering
020201 artificial intelligence & image processing
Artificial intelligence
business
media_common
Subjects
Details
- Database :
- OpenAIRE
- Journal :
- Proceedings of the Genetic and Evolutionary Computation Conference
- Accession number :
- edsair.doi...........76270ae6ca090b55c0470f340216f670
- Full Text :
- https://doi.org/10.1145/3449639.3459329