Talk to Algorithms to Live By: The Computer Science of Human Decisions like its author wrote you back.
Get the ideas that fit your life — not generic summaries.
- Chat with the book
- Audiobook-style main ideas
- Adapts to your life and goals
- Helps you take action
Algorithms to Live By: The Computer Science of Human Decisions, in detail
Brian Christian is a writer and Tom Griffiths is a cognitive scientist, and together they argue that computer science has worked out rigorous solutions to many of the problems humans face every day — when to stop searching for a better option, how to manage your schedule, how to sort your memory — and that these solutions are both interesting and useful. Algorithms to Live By, published in 2016, is an accessible account of those solutions and their implications for how we make decisions.
The book begins with the optimal stopping problem: when should you stop searching and commit to the best option you have found so far? The mathematical solution — look at 37 percent of your options without choosing, then pick the next option better than anything seen so far — gives you the best possible odds of choosing the best option overall. This applies to hiring, apartment hunting, and, in the book's famous framing, the secretary problem and romantic search. The 37 percent rule will not always get you the best outcome, but it is the strategy that maximizes the probability of success.
Subsequent chapters cover the explore-exploit tradeoff, sorting algorithms applied to filing and memory, caching applied to what to keep close at hand, Bayes' theorem applied to what to believe about uncertain events, and game theory applied to social situations. Each chapter connects a well-defined mathematical problem to a human decision context, shows the optimal computational solution, and then discusses what the solution implies.
The book is more nuanced than its premise might suggest. Christian and Griffiths consistently acknowledge that mathematical optimality and human optimality are not the same thing, that context matters, and that computational solutions often depend on assumptions about the structure of the problem that human life may not satisfy. The final chapter on computational kindness — structuring your requests and environments to reduce the cognitive load on others — is one of the more genuinely novel applications of the framework. Throughout, the writing is clear and the mathematics is handled gracefully without becoming inaccessible.
The big ideas
- 1.
The optimal stopping rule: look at 37 percent of your options without choosing, then select the next option that is better than anything seen so far. This maximizes the probability of finding the best option.
- 2.
The explore-exploit tradeoff: time spent exploring options has a cost — opportunity cost — that means the balance between trying new things and sticking with what works depends on how much time you have left.
- 3.
Sorting algorithms suggest that keeping things in roughly sorted order is often more efficient than perfect sorting — the cost of perfect organization exceeds the benefit for most real information sets.
