Algorithms to Live By

About the book

Algorithms to Live By takes the routines that computer science uses to sort, schedule, search, and stop, and sets them beside ordinary human choices. Brian Christian and Tom Griffiths show that many familiar dilemmas, from when to settle on an apartment to how long to keep looking for a parking space, have formal counterparts that researchers have studied for decades. The book moves through optimal stopping, explore versus exploit tradeoffs, sorting, caching, scheduling, and prediction, treating each as a problem with a known structure and known limits.

The authors write as an interdisciplinary pair, one from computer science and one from cognitive science, and the book reflects that pairing. They explain the mathematics in plain language, then test it against human behavior and practical constraints. A recurring theme is that optimal strategies often require information people do not have, so the useful lesson lies in recognizing which problem you face rather than applying a formula blindly. The result is a book about decision-making that takes computation seriously without treating computers as a model for how people should live.

The premise has appeal because it reframes familiar frustrations as instances of problems with names. Waiting too long or settling too early becomes a stopping problem. Constantly checking email becomes a scheduling question. The authors are careful to note where the analogy strains, which keeps the book from becoming a collection of clever but empty comparisons.

How to read it

Approach the book as a guided tour of problem types rather than a manual. Each topic gets its own explanation, and the value comes from learning to recognize the shape of a decision. Readers who want a single unified theory will not find one, and the authors do not promise one.

The prose is accessible and example-driven, with technical ideas introduced gradually. Some sections move faster than others, particularly where probability enters. Skim the formal passages on a first read and return to them when a specific decision makes the concept concrete. The book rewards patience more than speed.

Who it suits

Readers facing recurring decisions about timing, searching, and allocating attention will find useful vocabulary here. Managers, students choosing among options, and anyone who enjoys seeing everyday behavior explained through a formal lens will get the most from it. The book suits readers with curiosity about computer science who want an entry point that avoids code and prerequisites.

Reading considerations

The book explains models; it does not supply personal recommendations, and its advice should be read as description rather than instruction. Some analogies between machine algorithms and human life are looser than others, and readers should test each comparison against their own situation. The edition reflects research and examples available at that time, so later work in decision science may qualify specific claims.