Quantum-Inspired Neural Architecture Search
Authors:
- 1 July 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 70, 1-8
Abstract
Deep neural networks have gained attention in the last decade as significant progress has been made in a variety of tasks thanks to these new architectures. Most of the time, hand-designed networks are responsible for this incredible success. However, this engineering process demands considerable time and expert knowledge, which leads to an increasing interest in automating the design of deep architectures. Several new algorithms have been proposed to address the neural architecture search problem, but many of them require significant computational resources. Quantum-inspired evolutionary algorithms (QIEA) have their roots on quantum computing principles and present promising results in respect to faster convergence. In this work, we propose Q-NAS (Quantum-inspired Neural Architecture Search): a quantum-inspired algorithm to search for deep neural architectures by assembling substructures and optimizing some numerical hyperparameters. We present the first results applying Q-NAS on the CIFAR-10 dataset using only 20 K80 GPUs for about 50 hours. The obtained networks are relatively small (less than 20 layers) compared to other state-of-the-art models and achieve promising accuracies with considerably less computational cost than other NAS algorithms.Keywords
This publication has 14 references indexed in Scilit:
- Genetic CNN2017
- A genetic programming approach to designing convolutional neural network architectures2017
- Quantum-Inspired Features and Parameter Optimization of Spiking Neural Networks for a Case Study from AtmosphericProcedia Computer Science, 2015
- A new model for credit approval problems: A quantum-inspired neuro-evolutionary algorithm with binary-real representation2009
- Quantum-Inspired Evolutionary Algorithm: A Multimodel EDAIEEE Transactions on Evolutionary Computation, 2008
- Quantum-inspired genetic algorithms2002
- Quantum-inspired evolutionary algorithm for a class of combinatorial optimizationIEEE Transactions on Evolutionary Computation, 2002
- Evolving Neural Networks through Augmenting TopologiesEvolutionary Computation, 2002
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- An evolutionary algorithm that constructs recurrent neural networksIEEE Transactions on Neural Networks, 1994