Algorithms to Live By: How Computer Science Can Help You Make Better Decisions
In an age of information overload, decision-making is both more critical and more complex than ever. We are constantly faced with choices—some small, some life-changing—ranging from what to eat for lunch to where to invest our savings. How do we make the right decisions, minimize regret, and optimize our time and resources?
The field of computer science offers surprising insights into these everyday challenges. The book Algorithms to Live By by Brian Christian explores how concepts from computer science, particularly algorithms, can be applied to human decision-making. Just as algorithms help computers solve problems efficiently, they can help us make better, more informed choices in our personal and professional lives.
Here’s a look at some of the most important algorithms discussed in the book and how they can be applied to decision-making.
1. The Optimal Stopping Problem (The “37% Rule”)
One of the most famous algorithms that has real-world applications in human decision-making is the Optimal Stopping Problem, which deals with the question: When should you stop searching and make a decision? This is a critical question when you are faced with a series of options, but you don’t want to waste too much time, nor do you want to settle for something suboptimal.
For example, when you are house hunting, how many houses should you visit before deciding to make an offer? Or, when dating, how many people should you date before settling down with someone?
The Solution:
The optimal stopping algorithm tells us that you should spend a fixed proportion of your time or search on “exploration” (gathering data), and then stop exploring and make a decision based on what you have seen. Mathematically, the answer is 37%. In the case of house hunting, for example, you should spend the first 37% of your time (or, in the case of dating, the first 37% of potential partners) learning about your options. After that, you should choose the next option that is better than what you’ve already encountered.
This doesn’t guarantee that you will always make the best possible choice, but it significantly increases the odds that your decision will be close to optimal.
2. The Secretary Problem (A Variant of Optimal Stopping)
A variation of the optimal stopping problem is known as the Secretary Problem, where you are hiring a secretary (or making another choice) from a group of applicants or candidates. The question is: How do you decide when to select the best candidate, given that you can only evaluate candidates in sequence and must choose before you’ve seen them all?
The Solution:
The optimal strategy, as derived from probability theory, is to reject the first 37% of candidates and then hire the next candidate who is better than all those you’ve already seen. If no one beats the first 37%, you should choose the best candidate you’ve seen by the time you reach the end of the process. Again, this doesn’t guarantee that you will always select the best candidate, but it maximizes the chances of making the best choice.
This concept can be applied to various areas of life, including hiring employees, selecting business partners, or even deciding on investments.
3. The Explore-Exploit Tradeoff
The Explore-Exploit Tradeoff is a classic dilemma in decision theory: Should you explore new possibilities and try new things, or should you stick with what you know works and exploit it further? In many areas of life—whether it’s choosing a restaurant, managing investments, or learning new skills—this tradeoff plays a crucial role.
- Exploration involves trying out new options and gathering information.
- Exploitation involves taking advantage of what you already know works well.
The Solution:
The key is finding the right balance between exploration and exploitation. The optimal strategy often depends on the context, but one widely used approach is the ε-greedy algorithm, which says you should usually exploit what works best, but occasionally explore new options with a small probability (ε). The more time you have to gather information, the more you can afford to explore, but when you’re pressed for time or resources, you focus on exploitation.
This strategy applies not just in business or investing but also in personal growth, where sometimes trying new things (exploration) is necessary, but sticking with your strengths (exploitation) might be the best approach for long-term success.
4. Bayesian Decision Theory (Updating Beliefs in the Face of New Evidence)
Bayesian decision theory provides a framework for decision-making under uncertainty. It involves updating your beliefs based on new information or evidence. Rather than making a fixed decision based on initial assumptions, Bayesian decision-making suggests that we should continuously adjust our views and choices as we receive new data.
For example, if you are evaluating a potential business opportunity, you might start with an initial belief about how successful it could be. But as you gather more information—perhaps through customer feedback, market trends, or financial results—you adjust your expectations. In essence, the process of updating beliefs becomes a series of small decisions that refine your understanding of the world.
The Solution:
Bayesian decision theory uses probabilities to quantify uncertainty and Bayes’ Theorem to update these probabilities based on new evidence. The result is a more dynamic and flexible approach to decision-making that allows you to make better choices in uncertain environments.
This method can be applied to everything from stock market investing (adjusting your view of a company’s potential based on new financial data) to personal relationships (adjusting your assumptions about a partner based on new experiences or insights).
5. Game Theory (Cooperation vs. Competition)
Game theory is the study of strategic decision-making in situations where the outcome depends not only on your own decisions but also on the decisions of others. The classic example is the Prisoner’s Dilemma, where two individuals must choose whether to cooperate or betray each other, with varying consequences depending on the choices of both parties.
In the real world, game theory helps us understand interactions in situations like business negotiations, politics, and even everyday social dynamics. It teaches us when to cooperate, when to compete, and how to anticipate the behavior of others in competitive settings.
The Solution:
In many scenarios, cooperation leads to better overall outcomes for both parties than if everyone acts purely out of self-interest. Strategies like the Tit for Tat approach (where you cooperate initially and then mimic your opponent’s last move) can encourage mutual cooperation and lead to better long-term results than pure competition.
This strategy is valuable in business relationships, partnerships, and personal negotiations, where understanding the interplay of cooperation and competition can lead to win-win solutions.
6. The Takeaway: Algorithms as Decision-Making Tools
While computer science algorithms might seem like an abstract, technical subject, the principles behind them are surprisingly relevant to everyday life. By understanding and applying algorithms like the optimal stopping problem, the explore-exploit tradeoff, Bayesian decision theory, and game theory, we can make more informed and rational choices in areas such as career planning, investments, relationships, and personal development.
Conclusion: Using Algorithms to Improve Our Lives
Brian Christian’s Algorithms to Live By reminds us that decision-making doesn’t have to be random or based purely on gut feeling. Instead, we can use the insights from computer science to make smarter, more efficient choices. By applying algorithms designed to optimize decision-making, we can navigate the complexities of life with greater clarity, confidence, and success.
These algorithms don’t offer one-size-fits-all solutions but instead provide frameworks for thinking more strategically and making more informed choices. By understanding the underlying principles behind these algorithms, we can all become better at solving problems, managing uncertainty, and making decisions that improve our lives.