In the last year I actually ended up reading a few books (no, I don’t know how I did it either). Here I’m going to list out some important take away points from each book and what I think about the book overall. All of the books I read were non-fiction and somehow statistics related so they actually did have “lessons” in them. Here they go in order:

The Signal and the Noise - Nate Silver

This book has a good premise. Whenever humans try to make predictions or forecasts (they are different) they have to base them on some information. The difficult part of this information is that a lot of it is “noise” that doesn’t matter or just confuses us. What we need to look for is the “signal”.

The author makes a lot of cool case studies, such as how we have no idea how to accurately predict earthquakes but people try to (mistaking noise for signals). It also brings up the financial crisis and how many companies including the ratings agencies like moody’s saw it coming but chose to ignore it (ignoring the signal). It also brings up a cool story about how world champion Garry Kasparov lost in chess to IBM’s Deep Blue AI. He then talks about statistics for a while.

The author’s writing is good too but my problem with the book is that it doesn’t make clear how to tell signals apart from noise. In fact, all of the case studies make the comparison in hindsight. For me its a 3/5. I recommend it only if you are interested in this sort of stuff abd don’t know too much about it already.

Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy - Cathy O Neil

You would never think that big data and machine learning could actually have a negative impact on parts of society. That’s the big idea that this book explores. The case studies that are presented in this book are extremely eye opening and they highlight the problems of AI and machine learning in our society.

If an algorithm is based purely on a lot of data and math, it is probably right is one of the core ideas. The author notes that while machines help our society operate more efficiently, the businessmen that use the machines usually have absolutely no idea when they work. When people come to them to question the algorithm, they can’t change anything about the algorithms and the easy reply is to trust the algorithm. It can lead to a vicious cycle.

One eye opening case study the author points out is how targeted ads are used to pitch ads for for-profit colleges to poor neighborhoods. These colleges are about business and the degrees have been shown to be not worth much at all. Yet the targeted ads keep students going to them, and keeps the people in those neighborhoods in a vicious cycle of poverty. Another case study was with a crime prediction software used by the NYPD. Engineers fed decades of crime data into the machine learning algorithm to make it predict where the next crimes would most likely occur. It turns out this software keeps police patrolling the same low-income neighborhoods mentioned earlier. Moreoever since they patrol those neighborhoods more, they get more crimes there and the machine keeps telling them to go there. The worst part is no one can question a machine.

I highly recommend this one especially if you interested in ethics and big data. The writing was decent, the thing about this book that will keep you reading is the case studies. I give it a 3.5/5

What the Luck? The Surprising Role of Chance in Our Everyday Lives - Gary Smith

Despite being the book that taught me the least among the rest, I actually liked this one the most. The writing is very good and the way the author explains stuff makes a lot of sense.

The big thing that the author talks about here is called “Regression to the mean” and he gives a lot of cool examples. My favorite one is this: Smart women tend to date guys who are not as smart as them. It turns out that that is a fact. Now it’s easy to say that women like to feel intellectually superior or that guys are just not as smart in general but the author makes a key point. If you are a smart person, you are simply less likely to meet someone as smart as you. Everything in the book is based on this. He also talks about some other interesting things like standardized test grading and air force pilot training, but you can check it out yourself.

I definitely recommend this book, its worth a read if you want to learn more about how randomness plays a role in your everyday life. I give it a 4/5.

The Black Swan: The Impact of the Highly Improbable - Nassim Taleb

This book probably had the most knowledge packed into it, but it also had some of the most pretentious and hard to understand writing in it as well.

There is too much in this book to unpack but I’ll explain the title. It turns out that back in the day during the age of exploration people believed that all swans were white. They thought so cause they only ever saw white swans. Until the one day that they did see a black swan and it challenged everything they thought they know. This book is about those “black swan” events that no one sees coming. The author characterizes them, describes when they show up, and what kind of impact they could have.

If you’re looking for a good read don’t bother with this. If really you want to gain some knowledge about how uncertainty takes form and shapes events in the world, check this book out. I give it a 3.5/5.