Scan barcode
A review by matty_icey
Algorithms to Live by: The Computer Science of Human Decisions by Tom Griffiths, Brian Christian
4.0
This was an awesome book about the underlying algorithms in our world. Favorite parts:
Everyone rises to their own level of incompetence. Once you get to a level where you aren’t good at something, you don’t advance because you aren’t worthy of it, and thus are stuck. One can conclude that we are all doing what we are not great at.
On marriage: “If the prisoner is happy, why lock him in? If the prisoner isn’t, why pretend he is?” - George Shaw
On when you should stop looking for a better deal, parking spot, home, or job applicant: Stop after you’ve collected 30% of the data you want and then take the next one that is the best you’ve seen. Or take anything above your baseline average. This is scientifically your best odds at getting the best option. See the secretary dilemma.
Sometimes just relaxing one or two constraints can make things much easier. If you you want the perfect “whatever” let’s say vacation, then research planning etc is going to take time. But a good vacation may be just intuitive and take no time at all to plan without the hassle. Good is better than perfect.
You can beat computers in chess by playing the clock and after grid locking the board making random movements. The computer wastes time trying to figure out what your ever changing strategy is and loses by running out of time. This is assuming you can avoid getting check mated while making fast non-sensical moves. I loved this ha.
Everyone rises to their own level of incompetence. Once you get to a level where you aren’t good at something, you don’t advance because you aren’t worthy of it, and thus are stuck. One can conclude that we are all doing what we are not great at.
On marriage: “If the prisoner is happy, why lock him in? If the prisoner isn’t, why pretend he is?” - George Shaw
On when you should stop looking for a better deal, parking spot, home, or job applicant: Stop after you’ve collected 30% of the data you want and then take the next one that is the best you’ve seen. Or take anything above your baseline average. This is scientifically your best odds at getting the best option. See the secretary dilemma.
Sometimes just relaxing one or two constraints can make things much easier. If you you want the perfect “whatever” let’s say vacation, then research planning etc is going to take time. But a good vacation may be just intuitive and take no time at all to plan without the hassle. Good is better than perfect.
You can beat computers in chess by playing the clock and after grid locking the board making random movements. The computer wastes time trying to figure out what your ever changing strategy is and loses by running out of time. This is assuming you can avoid getting check mated while making fast non-sensical moves. I loved this ha.