| Title |
Straightforward Clarification for Fundamental Algorithms of Artificial Neural Networks |
| Authors |
(Nematullo Rahmatov) ; (Hoki Baek) |
| DOI |
https://doi.org/10.5573/IEIESPC.2023.12.3.223 |
| Keywords |
LSR; Loss function; Gradient descent; Backpropagation |
| Abstract |
Artificial neural networks (ANNs) have revolutionized the field of science in the last few decades. Unlike classical machine learning (ML) algorithms, which require human effort to craft well-structured features, an ANN automatically extracts complex patterns as features and passes them into ML to perform various downstream tasks, such as classification and segmentation. Hence, ANNs have made most classical ML algorithms obsolete for many tasks. In addition, deep learning-based models, such as convolutional neural networks, recurrent neural networks, graph neural networks, and generative adversarial neural networks, accelerate artificial intelligence (AI) applications. Therefore, it is essential for novices in ML to understand the basic functionality of ANN to pursue deep learning-related algorithms. Considering this importance, this paper explains the major functionalities of ANN algorithms, such as loss function and backpropagation. |