Modern AI tools are powerful, but they can fail in unexpected ways. This often happens because we do not see the hidden ideas and assumptions behind them. To build better systems, we need to understand the basics on which they are built. This book connects the AI tools we use every day to the original ideas behind them.
- Balance model accuracy with fairness and transparency in the systems.
- Learn to think in probabilities by using ideas like Bayesian inference and entropy.
- Design trustworthy systems by making careful choices about data and model structure.
- Recognise the hidden assumptions built into every model.
- Use automation tools like generative AI and AutoML with human oversight.
Timeless Algorithms helps you understand the key ideas from early AI research. It uses these ideas to help you find problems and understand how different parts of a model work. Each chapter links a common data tool to the original mathematics paper that made it possible.
You will learn how to design systems you can trust. The book helps you think using probability and use automation in a careful way. This will help you build AI systems that are both effective and responsible. This book is for data scientists, engineers, statisticians and business analysts. It is also useful for anyone who makes decisions using AI.
Table of Contents:
1 SEEING INSIDE THE BLACK BOX
2 FROM EFFECT TO CAUSE: BAYES' THEOREM AND THE FIRST ALGORITHM OF LEARNING
3 THE ALGORITHM OF ESTIMATION: RONALD FISHER'S LIKELIHOOD PRINCIPLE
4 TESTING WHAT WE ASSUME TO KNOW: NEYMAN, PEARSON, AND THE PRINCIPLES OF HYPOTHESIS TESTING
5 THE BIRTH OF INFORMATION THEORY: SHANNON AND THE MATHEMATICS OF UNCERTAINTY
6 THE LOGIC OF MULTI-STAGE DECISION PROCESSES: RICHARD BELLMAN AND THE PRINCIPLE OF RECURSIVE OPTIMIZATION
7 FROM INFERENCE TO CHOICE: HOWARD RAIFFA, ROBERT SCHLAIFER, AND THE BAYESIAN REVOLUTION
8 THE GEOMETRY OF SEPARATION: VLADIMIR VAPNIK AND THE MATHEMATICS OF SUPPORT VECTOR MACHINES
9 FROM SINGLE TREES TO FORESTS: LEO BREIMAN AND THE LOGIC OF ENSEMBLE LEARNING
10 FROM ISOLATED ALGORITHMS TO COHERENT SYSTEMS: DAVID J.C. MACKAY AND THE UNIFYING LOGIC OF LEARNING
11 LEARNING THROUGH REPRESENTATION: LECUN, BENGIO, HINTON, AND THE MATHEMATICS OF NEURAL NETWORKS
12 FROM RECURRENCE TO ATTENTION: GOOGLE BRAIN AND THE TRANSFORMER ARCHITECTURE
13 FROM FOUNDATIONS TO FRONTIER: OPEN AI AND THE SCALING LAWS OF MODERN INTELLIGENCE
About the Author :
Gary Sutton is a leader in business intelligence and analytics. He is known for making complex statistical ideas easy to understand. With his experience in the field, he brings a practical and clear voice to his writing. Gary Sutton helps readers connect timeless theories to the real-world challenges of building modern AI systems.
