Neural Network for handwritten digit classification with different training approach
- DOI
- 10.2991/978-94-6463-738-0_16How to use a DOI?
- Keywords
- Convolution Neural-Network; NumPy; Backpropagation; Data Sets; MNIST; Perceptron
- Abstract
Artificial Neural Network is more interesting than what it just seems to be. The new world of inventions and innovation, all the possibilities opened up due to a simple arrangement of neurons in layers is stunning. The scope even widens up when the internal structure of the ANN or CNN is untangled. The neural network designed here from scratch displayed its accuracy on training data to be 99.87% and 98.88% on testing data. A deeper understanding of the neural network, its structure and its working enables one to formulate and design a neural network according to the requirements, which is here applied to its training process. Recognising handwritten digits implemented here is one of the simplest and most effective tasks a neural network can be designed for, so as to gain the understanding of the internals of the NN.
- Copyright
- © 2025 The Author(s)
- Open Access
- Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
Cite this article
TY - CONF AU - Manav Raj Siddhant Kunvar AU - Ankita Sharma PY - 2025 DA - 2025/06/22 TI - Neural Network for handwritten digit classification with different training approach BT - Proceedings of the International Conference on Advances and Applications in Artificial Intelligence (ICAAAI 2025) PB - Atlantis Press SP - 192 EP - 207 SN - 1951-6851 UR - https://doi.org/10.2991/978-94-6463-738-0_16 DO - 10.2991/978-94-6463-738-0_16 ID - Kunvar2025 ER -