Abstract
Convolutional neural network is the deep learning model which has several hidden layers in contrast to feed-forward neural network. Modeling of CNN layers depends upon the dataset and very challenging task as several trials are required to select the CNN parameters. In our work, we presented an optimal solution to tune the hyperparameters of CNN architecture by clonal search algorithm (CSA). This is tested on a challenging dataset of EMNIST, which is enhanced from ML benchmark dataset of NIST. With the proposed algorithm, it is possible to get the accuracy up to 98.7%.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Grother P (1995) NIST special database 19 handprinted forms and characters database. National Institute of Standards and Technology
Muthreja I, Kaur D (2018) A comparative analysis of immune system inspired algorithms for traveling salesman problem. In: International conference on artificial intelligence, pp 164–170
de Castro LN, Von Zuben FJ (2002) Learning and optimization using the clonal selection principle. IEEE Trans Evol Comput 6(3):239–251
Yim J, Ju J, Jung H, Kim J (2015) Image classification using convolutional neural networks with multi-stage feature. In: Advances in intelligent systems and computing, vol 345. Springer, Cham
Jaswal D, Sowmya V, Soman KP (2014) Image classification using the convolutional neural network. Int J Advancements Res Technol 3(6)
Xu L, Ren JS, Liu C, Jia J (2014) Deep convolutional neural network for image deconvolution. In: 27th international conference on a neural information processing system, pp 1790–1798
Nakahara H, Yonekawa H, Sato S (2017) An object detector based on multiscale sliding window search using a fully pipelined binarized CNN on an FPGA. In: 2017 international conference on field-programmable technology (ICFPT), Melbourne, VIC, pp 168–175
Bhandare A, Kaur D (2018) Designing convolutional neural network architecture using a genetic algorithm. In: International conference of artificial intelligence, pp 150–156
Bappy JH, Roy-Chowdhury AK (2016) CNN based region proposals for efficient object detection. In: IEEE international conference on image processing (ICIP), Phoenix, AZ, pp 3658–3662
Farooq A, Anwar S, Awais M, Rehman S (2017) A deep CNN based multi-class classification of Alzheimer’s disease using MRI. In: IEEE international conference on imaging systems and techniques (IST), Beijing, pp 1–6
Le Cun Y et al (1989) Handwritten digit recognition: applications of neural network chips and automatic learning. IEEE Commun Mag 27(11):41–46
Zhao R, Song W, Zhang W, Xing T, Lin J-H, Srivastava M, Gupta R, Zhang Z (2017) Accelerating binarized convolutional neural networks with software-programmable FPGAs. In: ISFPGA, pp 1–10
Cohen G, Afshar S, Tapson J, van Schaik A (2017) EMNIST: an extension of MNIST to handwritten letters, pp 1–10
Rastegari M, Ordonez V, Redmon J, Farhadi A (2016) XNOR-Net: ImageNet classification using binary convolutional neural networks, pp 1–55
Ren S, He K, Girshick R, Sun J (2017) Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans Pattern Anal Mach Intell 39(6):1137–1149
Jmour N, Zayen S, Abdelkrim A (2018) Convolutional neural networks for image classification. In: 2018 international conference on advanced systems and electric technologies (IC_ASET), Hammamet, pp 397–402
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Bohra, N., Bhatnagar, V. (2021). Tuning of CNN Architecture by CSA for EMNIST Data. In: Goar, V., Kuri, M., Kumar, R., Senjyu, T. (eds) Advances in Information Communication Technology and Computing. Lecture Notes in Networks and Systems, vol 135. Springer, Singapore. https://doi.org/10.1007/978-981-15-5421-6_6
Download citation
DOI: https://doi.org/10.1007/978-981-15-5421-6_6
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-15-5420-9
Online ISBN: 978-981-15-5421-6
eBook Packages: EngineeringEngineering (R0)