AI Algorithms
Albert J. Marcella · 2025
Chapter 2 examines the types and functionalities of AI algorithms, including supervised, unsupervised, reinforcement, and hybrid learning methods. It emphasizes the role of data in training algorithms, stressing accuracy and bias mitigation. Practical applications such as anomaly detection, customer segmentation, and autonomous systems are explored. The chapter also introduces advanced techniques like generative adversarial networks (GANs) and model-based reinforcement learning alongside audit guidelines for ensuring ethical and robust AI deployments. The importance of algorithm audits to improve transparency, equity, and accountability in AI systems is underscored. The chapter provides audit questions to evaluate the design, functionality, and ethical implications of supervised unsupervised, reinforcement, and hybrid AI algorithms.