Image Classification Based on Automatic Neural Architecture Search Using Binary Crow Search Algorithm

Abstract

Neural architectures have accelerated the advancement in various domains by enabling automatic pattern detection, image classification, audio recognition, and face recognition etc. However, they are computationally expensive to design and expert knowledge in various domains is required. In this paper, a swarm intelligence algorithm is proposed to search for novel architectures without human intervention that can achieve comparable performance to those of human-designed architectures. This work is inspired by current neural architecture search approaches based on reinforcement learning and genetic algorithm. However, not much attention is paid towards swarm intelligence metaheuristics-based neural architecture search. A framework is proposed for automatically designing neural architectures based on a swarm intelligence metaheuristic: Crow Search Algorithm. First, Crow Search Algorithm is integrated with binary network representation. To make it compatible for Neural Architecture Search, the original distance metric is replaced with hamming distance-based similarity measure. Second, the tuning parameters of Crow Search Algorithm are reduced by replacing the static flight length parameter with our dynamic flight length distribution algorithm. Third, the target selection method (random selection) is replaced by tournament select method. The proposed framework is used to search for architectures on MNIST, CIFAR10, and CIFAR100 datasets and achieved 0.18%, 3.48%, and 15.64% test error, respectively. Furthermore, smallscale transfer experiments are conducted to search architectures for Tiny ImageNet and achieved 34.43% test error. Nonparametric statistical analysis is performed to validate the impact of each modification in improving the quality of search space exploration. The proposed framework has achieved comparable performance with the state-of-the-art approaches, with a comparatively simpler approach and minimum human intervention. The proposed framework can be used to develop completely automated systems for designing architectures for various data-based classification applications.

A survey on binary metaheuristic algorithms and their engineering applications

“…Ahmad et al proposed a swarm intelligence algorithm to search for new architectures without manual intervention (Ahmad et al. 2020 ). Firstly, CSA is combined with binary network representation, and hamming distance-based similarity replaces original distance metric to make it compatible with neural structure.…”

“…( 2020 ) CSO & Salp Swarm Algorithm CNN Medium High Low Ahmad et al. ( 2020 ) Binary CSA CNN Low High Low Zhang et al. ( 2019a ) DE & Binary DE NN Medium High Low Li et al.…”

Intelligent Military Aircraft Recognition and Identification to Support Military Personnel on the Air Observation Operation

“…The distance between two binary strings is the total amount of different bits at the same position [45]. We select Hamming Distance for our system based on its good performance application for recognition as well as for decoding application where similarity is used as the primary measurement for the system or method performance [46], [47], [48], [49].…”

Improved Crow Search Algorithm Optimized Extreme Learning Machine Based on Classification Algorithm and Application

“…research results in the fields of network optimization and distribution, medical testing and so on [30]. The crow search algorithm has only two parameters (flight length and perception probability), and the crow search algorithm is easy to implement and its convergence speed is fast.…”