Artificial Intelligence: a Mathematical and Empirical Exploration

EasyChair Preprint 15724

Abstract

Artificial Intelligence (AI) has rapidly evolved over the past few decades, revolutionizing a wide range of industries including healthcare, finance, transportation, and entertainment. This paper provides an in-depth exploration of AI, with a focus on both its mathematical foundations and empirical applications. We begin by discussing the core mathematical models that drive AI systems, including artificial neural networks (ANNs), machine learning (ML) algorithms, and optimization techniques. We emphasize how these models are formulated mathematically, explaining key equations and their relevance in real-world scenarios.

Furthermore, this paper presents a comprehensive comparison of several AI algorithms through a series of experiments designed to evaluate their effectiveness in solving practical problems. These experiments involve real-world datasets in domains such as financial forecasting and medical diagnostics, with a particular focus on prediction accuracy and computational efficiency. The results are presented in both tabular and graphical formats, allowing for a clear analysis of the performance trade-offs between different models.

The findings indicate that while more complex models, such as deep learning-based artificial neural networks, tend to achieve higher accuracy rates, they require more computational resources and longer execution times. Conversely, simpler models like decision trees and support vector machines offer faster processing times but may compromise on prediction accuracy. This trade-off between accuracy and efficiency is a central challenge in the application of AI, and the choice of model often depends on the specific constraints and goals of the problem at hand.

Keyphrases: ANN, Artificial Intelligence, Finance, mathematical

@booklet{EasyChair:15724,
  author    = {Rene Gozalens and Isabel Cheng and Amir Samavati and Che Hoo and James Kung},
  title     = {Artificial Intelligence: a Mathematical and Empirical Exploration},
  howpublished = {EasyChair Preprint 15724},
  year      = {EasyChair, 2025}}