Book Review #30
Recently I came across a book reading rule that said calculate 100 minus your age and if by that page number you are not excited drop the book. Meaning if you’re 100, you’re allowed to judge a book by its cover 😀
I started 'Algorithms to Live by' thinking of this rule but within the first few pages was convinced to finish the book. Brian Christian and Tom Griffiths explain several useful Computer Algorithms in plain English and helps us with valuable lessons on how to apply them in our everyday life. Here are my top takeaways from the book
1> The 37% rule: Imagine you are apartment hunting, spend 37% of your time (11 days if you have a month for the search) noncommittally exploring options. But after that point, immediately commit—deposit to the very first place that beats whatever you’ve already seen. You can use this for hiring candidates or even dating to find your spouse.
2> TCP/AIMD: Imagine you hired a new person, give them first a small piece of work and if they succeed, give them double the amount of work. Keep doubling the load till they fail and that point halve the load and slowly increase step by step. This is the approach of Internet's Transmission Control Protocol (TCP) and the Additive Increase, Multiplicative Decrease (AIMD) approach helps you quickly understand the capacity of the new person
3> Pivot or Persevere using Gittens index: The Gittens Index is a mathematically proven approach where you can compare the success probability of your current approach vs. an alternate new approach to decide whether to continue (persevere) to invest or move to a new option (pivot). Thumb rule - if your current approach has less than 70% success rate and a new untested approach has similar potential you should pivot to the new approach.
4> Sorting: Sort anything you have only if you will have a need to search later as sorting theory shows "Scale hurts" and if the cost of searching drops, sorting becomes less valuable. Think before you sort/organize all your email vs. the ones you will need to retrieve.
5> Caching: In the practical use of our intellect, forgetting is as important a function as remembering. Like a computer's cache (temporary memory), the more you try to remember, the worse off you are. Simple rule - our best guide to the future is a mirror image of the past. The nearest thing to clairvoyance is to assume that history repeats itself.
6> Small Data: Many daily decisions need to be made with minimal data (e.g. should I continue to wait in the bus stand or take a taxi). Having a small amount of prior data (minutes since the last bus left) can help us make extremely good predictions (Read more on Bayes theorem if interested)
The book also has many more interesting sections on Explore vs. Exploit tradeoffs, Recursion, Optimal stopping and Game theory shaping human behavior. Overall a great read !
#bookreview #decisionmaking
No comments:
Post a Comment