I have been studying **Artificial Intelligence** since 2017. That is four years. And I have learned nothing. Well, that is not quite true. I have gotten pretty familiar with the terminology. However, after fours years of intense effort I still don’t know how to do anything with machine learning. I can install the libraries and run the demos or the code in tutorials, but when it comes time to do something with original data I don’t have a clue.

Unfortunately, most tutorials and articles on machine learning are poorly written. They don’t explain the math or the theory. Often they don’t provide more than one example. The example given is usually a toy problem using a classic data set. For example, I have eleven examples in my notes using the Iris Flower data set. To be fair, my study methods are part of the problem. I randomly search for tutorials and copy and paste the code to run the demo.

Fortunately, I have begun to make some real progress by concentrating on a single algorithm or method. I have mastered **Linear Regression** because it is widely used in statistics. Tutorials on statistics explain linear regression in far greater detail. I can now perform a linear regression in Excel, C#, Python, R Studio, and even Processing (JavaScript). I have also mastered **Multiple Regression** since that is just Linear Regression using more independent variables. I am poised to master simple **Neural Networks** since I have successfully used a Python implementation from scratch with multiple data sets.

So far I have read the following books on Artificial Intelligence:

**Our Final Invention: Artificial Intelligence and the End of the Human Era**James Barrat**Thinking Machines: The Quest for Artificial Intelligence and Where It’s Taking Us Next**Luke Dormehl**Machine Learning**Ethem Alpaydin**Apocalyptic AI: Visions of Heaven in Robotics, Artificial Intelligence, and Virtual Reality**Robert M. Geraci**Artificial Intelligence: The Quest for the Ultimate Thinking Machin**e Richard Urwin**Natural Language Processing with Python: Analyzing Text with the Natural Language Toolkit**Steven Bird, Ewan Klein, Edward Loper**Text Analytics With Python**Dipanjan Sarkar**Bayesian Statistics the Fun Wa**y Will Kurt**AIQ: How Artificial Intelligence Works and How We Can Harness Its Power for a Better World**Nick Polson, James Scott**Superintelligence: Paths, Dangers, Strategies**Nick Bostrom**Introducing Artificial Intelligence: A Graphic Guide**Henry Brighton, Howard Selina**Statistical Inference via Data Science: A ModernDive into R and the Tidyverse**Albert Y. Kim, Chester Ismay**The Master Algorithm: How the Quest for the Ultimate Learning Machine Will Remake Our World**Pedro Domingos**The Quest for Artificial Intelligence**Nils J. Nilsson

Currently I am reading *Statistics In Plain English* by Timothy C. Urdan because you really need to understand statistics to grasp machine learning. I am also reading *Practical Machine Learning in R* by Fred Nwanganga and Mike Chapple which should prove easy to understand since it does not include much math or theory.

I am doing some free online courses at Kaggle.

Ordinarily I would give up on machine learning because it is just too difficult to learn and it does not appear to have any practical uses for me. But I am fascinated by the topic of artificial intelligence so I keep plugging away. I will say that I have vastly improved my knowledge of math, statistics, graph theory, linear algebra, calculus, combinatorics, and various other computer science topics. I am now more familiar with mathematical notation. I use MathJax to include math equations in my notes which I keep in HTML.

I have not bought much hardware to support my study of artificial intelligence. I have a Jetson Nano and a Neural Compute Stick 2. Eventually I might buy a NVIDIA GPU.