Regularized Evolutionary Algorithm for Dynamic Neural Topology Search
| Autor: | Nicu Sebe, Subhankar Roy, Giovanni Iacca, Cristiano Saltori |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Network architecture
Artificial neural network business.industry Computer science Deep learning Cognitive neuroscience of visual object recognition Process (computing) Evolutionary algorithm 02 engineering and technology 010501 environmental sciences 01 natural sciences 0202 electrical engineering electronic engineering information engineering Memory footprint 020201 artificial intelligence & image processing Artificial intelligence business MNIST database 0105 earth and related environmental sciences |
| Zdroj: | Lecture Notes in Computer Science ISBN: 9783030306410 ICIAP (1) |
| Popis: | Designing neural networks for object recognition requires considerable architecture engineering. As a remedy, neuro-evolutionary network architecture search, which automatically searches for optimal network architectures using evolutionary algorithms, has recently become very popular. Although very effective, evolutionary algorithms rely heavily on having a large population of individuals (i.e., network architectures) and are therefore memory expensive. In this work, we propose a Regularized Evolutionary Algorithm with low memory footprint to evolve a dynamic image classifier. In details, we introduce novel custom operators that regularize the evolutionary process of a micro-population of 10 individuals. We conduct experiments on three different digits datasets (MNIST, USPS, SVHN) and show that our evolutionary method obtains competitive results with the current state-of-the-art. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|