Predicting the Unknown: The History and Future of Data Science and Artificial Intelligence

ISBN-13: 9781484295045 / Angielski

Stylianos Kampakis

Spis treści

Preface

Author’s note to the curious reader

Prologue

Chapter One – Where are we now? A brief history of uncertainty

Not all uncertainty is created equal

Chapter Two - Truth, logic and the problem of induction

The first black swan

Chapter Three - Swans and Space Invaders

Occam’s razor, space invaders and lizard people

Chapter Four - Probability: to Bayes, or not to Bayes?

Frequentist or Bayesian?

The formulation of Bayes’ theorem

After Laplace

Chapter Five - What’s Maths Got to do with it? The Power of Probability Distributions

Other Distribution Models

Issues with this view of uncertainty

Bounds and limits

Chapter Six - Alternative Ideas: Fuzzy Logic and Information Theory

Information Theory – Measuring Uncertainty

Chapter Seven – Statistics: the Oldest Kid on the Block

Descriptive vs Inferential Statistics

Hypothesis Testing: Significant or Not?

What the p?

Statistical modelling: A useful abstraction

Chapter Eight - Machine Learning: Inside the Black Box

Data Science and History of Machine Learning

Choose Your Learning Type: Supervised, Unsupervised, Reinforcement, or Other?

The Bias-Variance Trade-Off

Machine Learning vs Statistics: Why the ‘Dumb’ Approach Works

Machine Learning Shortcomings

Chapter Nine - Causality: Understanding the ‘Why’

How to Approach Causality?

Causality in our Mind

Chapter Ten - Forecasting, and Predicting the Future: The Fox and the Trump

A brief history of forecasting

Forecasting in practice: Newton and the madness of men, Trump, Brexit, and losing money through mathematical modelling

How to do forecasting: A story of foxes and hedgehogs

Chapter Eleven - The Limits of Prediction (part A): A futile Pursuit?

Learning theory: what can we know about what we don’t know?

Monte Carlo Simulations: What Does a Casino Have to do with Science?

Chapter Twelve - The Limits of Prediction (Part B): Game Theory, Agent-based Modelling and Complexity (Actions and Reactions)

Agent-based Modelling: Crafting artificial Worlds

Complexity Theory: Simulation vs the Limits of Prediction

Studying Complexity is a Complex Endeavour

Learning from Complexity: The Limits of Computation are the Limits of Uncertainty

Chapter Thirteen - Uncertainty in Us: How the Human Mind Handles Uncertainty

Uncertainty and our Mind

Uncertainty and our Brain

Chapter Fourteen - Blockchain: Uncertainty in transactions

The Internet of Trust

How Blockchain Works

From Crypto-Anarchism to Drug Trafficking: The unconventional Beginnings of an interesting Technology

I Can’t Trust you, but I Can Trust the Blockchain

Chapter Fifteen - Economies of Prediction: A New Industrial Revolution

Uncertainty brokers

Industries of incomplete Information

Prediction Industries and Automation

The global Economy against Uncertainty

Epilogue: The Certainty of Uncertainty

O autorze

Dr. Stylianos (Stelios) Kampakis is a data scientist, data science educator and blockchain expert with more than 10 years of experience. He has worked with decision makers from companies of all sizes: from startups to organizations like the US Navy, Vodafone ad British Land. His work expands multiple sectors including fintech (fraud detection and valuation models), sports analytics, health-tech, general AI, medical statistics, predictive maintenance and others. He has worked with many different types of technologies, from statistical models, to deep learning to blockchain and he has two patents pending to his name. He has also helped many people follow a career in data science and technology.

He is a member of the Royal Statistical Society, honorary research fellow at the UCL Centre for Blockchain Technologies, a data science advisor for London Business School, and CEO of The Tesseract Academy and tokenomics auditor at Hacken. As a well-known data-science educator, he has published two books, both of them getting 5 stars on Amazon. His personal website gets more than 10k visitors per month, and he is also a data science influencer on LinkedIn.

Blurb

As a society, we’re in a constant struggle to control uncertainty and predict the unknown. Quite often, we think of scientific fields and theories as being separate from each other. But a more careful investigation can uncover the common thread that ties many of those together. From ChatGPT, to Amazon’s Alexa, to Apple’s Siri, data science, and computer science have become part of our lives. In the meantime, the demand for data scientists has grown, as the field has been increasingly called the “sexiest profession.” 

This book attempts to specifically cover this gap in literature between data science, machine learning and artificial intelligence (AI). How was uncertainty approached historically, and how has it evolved since? What schools of thought exist in philosophy, mathematics, and engineering, and what role did they play in the development of data science? It uses the history of data science as a stepping stone to explain what the future might hold. 

Predicting the Unknown provides the framework that will help you understand where AI is headed, and how to best prepare for the world that’s coming in the next few years, both as a society and within a business. It is not technical and avoids equations or technical explanations, yet is written for the intellectually curious reader, and the technical expert interested in the historical details that can help contextualize how we got here. 

You will:

• Explore the bigger picture of data science and see how to best anticipate future changes in that field
• Understand machine learning, AI, and data science
• Examine data science and AI through engaging historical and human-centric narratives