When I first ordered this book through Amazon, I literally got an audio-book in a compact disk (CD) instead of the cover book which I paid like $30 dollars (CAD). I did not bother to bring this up to the seller as it was my fault for not properly selecting the option on the site. I ordered the book again only this time I made it sure the option is properly selected to hardcover. As soon as I received around last month, I started reading this book, “Algorithms to live by”. I finally finished it this week. The book is interesting and kept me reading until the end. I wouldn’t say I enjoyed book’s every detail, but it is well researched and provides good examples of daily human decisions from computer science point of view. The interesting part is that the authors do not necessarily have hard-core computing science background which is probably why the book kept me interesting . I did not find reading this book much boring but I found myself eager to find out what’s covered in next chapter. Every chapter is well constructed, written, and informative. The book basically shows us how to efficiently solve problems in our everyday lives.
The thing I really liked about this book is consistent examples provided in most of the chapters around the points authors making. You will find abundant examples throughout the book.
I particularly enjoyed when author covers regret minimization since it aligns with my life motto.
Also, “less is more” meaning it is often rational to ignore information in making decisions to prevent “overfitting”. What is overfitting? we easily assume more is always better, which might lead us to have better pros/cons list, better decisions, etc. However, this isn’t necessarily going to get us the best prediction. It is indeed true that including more factors in a model will always, by definition, make it a better fit for the data we have already (the past stuff). However, a better fit for the available data does not necessarily mean a better prediction for the future. It’s not always better to use the complex model.
The book also discusses on Bayes’s Rule which is used to measure the conditional probability of an event happening given another event happening is true. This can be written in the equation of where A and B are events and P(B) ≠ 0.
These are just 3 points where I found myself fascinating reading the book. There are obviously a lot more examples and points that caught my attention. It could literally change your mind of how you are thinking of in various situations. I highly recommend this book.
My Overall Rating: