Algorithms to Live By by Brian Christian and Tom Griffiths is a fascinating bridge between the rigid logic of computer science and the messy reality of human life. The core premise is that the issues we face—like choosing a spouse, organizing a closet, or managing a busy schedule—are actually problems of limited space and time, which computer scientists have already spent decades solving.
Here are the most impactful "algorithms" from the book and how they apply to your daily life:
1. Optimal Stopping: The 37% Rule
When you are searching for something (an apartment, a job, or a life partner) and you don’t know what options will appear in the future, when do you stop looking and commit?
The Math: To maximize your chances of picking the absolute best option, you should spend the first 37% of your search time (or 37% of the total options) simply "looking" without committing.
The Strategy: After that "look" phase, you should immediately commit to the very next option that is better than everything you’ve seen so far.
Application: If you give yourself a month to find a new apartment, gather data for 11 days. On day 12, sign for the first place that beats the best one you saw during those 11 days.
2. Explore vs. Exploit: The Multi-Armed Bandit
Should you go to your favorite restaurant (exploit) or try that new place down the street (explore)?
The Logic: Your decision should be based on time horizon. If you are in a city for only one night, "exploit" your favorite spot. If you just moved there, "explore" to find new favorites.
The Insight: We often view "exploring" as a failure if the experience is bad, but algorithmically, a "bad" meal is actually valuable data that narrows your future search.
3. Sorting and Searching: The Cost of Order
We spend a lot of time organizing things, but computer science asks: Is the cost of sorting worth the time saved searching?
The Strategy: If you rarely search for something, don't bother sorting it.
The "Least Recently Used" (LRU) Principle: Computers keep the most recently used data in "cache" (the most accessible spot). You can do the same with your closet: when you wash a shirt, hang it on the left side. Over time, the clothes you actually wear migrate to the left, and the ones you don't stay on the right.
4. Scheduling: Beating Procrastination
How do you handle a massive to-do list? Computers use different "heuristics" depending on the goal:
To reduce anxiety: Use Shortest Processing Time. Knock out the 2-minute tasks first. This reduces the number of items on your plate quickly.
To meet deadlines: Use the Earliest Due Date algorithm. Focus strictly on what is due next.
When you’re stuck: Use Preemption. Switch tasks when you hit a wall, but be careful—switching tasks has a "context switching" cost (the time it takes for your brain to "reload" the new task).
Why It Matters
The book's most comforting takeaway is that perfection is often computationally impossible. Sometimes, the "rational" thing to do is to use a "heuristic"—a mental shortcut—or even to embrace randomness when you’re stuck in a loop.
"Even the best strategy doesn’t guarantee a perfect outcome, just the best statistical chance of success."
Would you like me to dive deeper into one of these specific algorithms, or perhaps help you apply one of them to a decision you’re currently facing?
0 Comments