Introduction to Neural Networks and Deep Learning
Dimitrios Xanthidis, Muhammad Fahim, Han‐I Wang · 2022
Deep learning (DL) is a specific form of ML, and therefore another branch of AI. At a basic level, DL is based on mimicking the human thinking process and developing relevant abstractions and connections. It consists of the following elements: (a) Learning: Facilitating the functionality to artificially obtain and process new information, (b) Reasoning: Offering the functionality to process information in different, and potentially overlooked, ways, (c) Understanding: Providing ways to showcase the results of the adopted model, (d) Validating: Offering the opportunity to validate the results of the model based on theory, (e) Discovering: Providing the mechanisms to identify new relationships within the data, (f) Extracting: Allowing the extraction of new meanings based on the predictors. This chapter provides an introduction to the basic mathematics and other concepts related to the Neural Networks and to the basics of DL as suggested above.