Evolutionary design of deep neural networks
| Autor: | Baldominos Gómez, Alejandro |
|---|---|
| Přispěvatelé: | Sáez Achaerandio, Yago, Isasi, Pedro, UC3M. Departamento de Informática, Ministerio de Educación, Cultura y Deporte (España), Universidad Carlos III de Madrid. Departamento de Informática |
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: | |
| Zdroj: | e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid instname |
| Popis: | Mención Internacional en el título de doctor For three decades, neuroevolution has applied evolutionary computation to the optimization of the topology of artificial neural networks, with most works focusing on very simple architectures. However, times have changed, and nowadays convolutional neural networks are the industry and academia standard for solving a variety of problems, many of which remained unsolved before the discovery of this kind of networks. Convolutional neural networks involve complex topologies, and the manual design of these topologies for solving a problem at hand is expensive and inefficient. In this thesis, our aim is to use neuroevolution in order to evolve the architecture of convolutional neural networks. To do so, we have decided to try two different techniques: genetic algorithms and grammatical evolution. We have implemented a niching scheme for preserving the genetic diversity, in order to ease the construction of ensembles of neural networks. These techniques have been validated against the MNIST database for handwritten digit recognition, achieving a test error rate of 0.28%, and the OPPORTUNITY data set for human activity recognition, attaining an F1 score of 0.9275. Both results have proven very competitive when compared with the state of the art. Also, in all cases, ensembles have proven to perform better than individual models. Later, the topologies learned for MNIST were tested on EMNIST, a database recently introduced in 2017, which includes more samples and a set of letters for character recognition. Results have shown that the topologies optimized for MNIST perform well on EMNIST, proving that architectures can be reused across domains with similar characteristics. In summary, neuroevolution is an effective approach for automatically designing topologies for convolutional neural networks. However, it still remains as an unexplored field due to hardware limitations. Current advances, however, should constitute the fuel that empowers the emergence of this field, and further research should start as of today. This Ph.D. dissertation has been partially supported by the Spanish Ministry of Education, Culture and Sports under FPU fellowship with identifier FPU13/03917. This research stay has been partially co-funded by the Spanish Ministry of Education, Culture and Sports under FPU short stay grant with identifier EST15/00260. Programa Oficial de Doctorado en Ciencia y Tecnología Informática Presidente: María Araceli Sanchís de Miguel.- Secretario: Francisco Javier Segovia Pérez.- Vocal: Simon Lucas |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|