As you probably know, every day that passes comes with its share of choices and decisions to be made. Which restaurant to choose? Which way to get there on time? How to answer the last email you received?…
And, as with all humans on this planet, making a decision is always based on the same simple, cold and relentless logic : We always want to make the right choice.
Over the course of your life, there will be millions of choices you could make and billions of subtle variations. And, considering that every decision you make affects, to a greater or lesser extent, the myriad of decisions you will have to make, the issue is of paramount importance.
Of course, some choices are easier than others. Eating an ice cream or a pancake is certainly a simpler decision than changing or not changing career paths, marrying someone or buying a house.
However, the underlying logic is the same: We make choices to “get what we want”.
What if “getting what we want” changes over time? So what do we do then?
Unless there is some form of omniscience, is there a better strategy than blind guessing and hoping for the best?
The answer is “yes”.
The problem of the bandit
Determining which decisions will give you the best results is very similar to a very important problem in probability theory: the problem of the multiarmed bandit.
Imagine entering a casino and deciding to play slot machines.
There is a row of machines, each of which has a different probability of paying a reward when you pull the lever. Some machines pay more — others much more — than others, but you are not sure which machine has the highest earning expectation.
👁️🗨️👁️🗨️ Also read : Why your innovation may go off track without a clear strategy
If you knew the best machine in advance, you’d just have to pull that lever all day long, but you have no idea, and no one will tell you. The only way to know is to start pulling the levers, to be attentive, to follow what works and what doesn’t, and to do the math.
However, there is a trade-off: Every time you choose to pull a lever that you have never pulled before, you get new and valuable information to find the best machine in the casino. But pulling on the least tested lever has a cost (of opportunity) since you are not pulling on the lever you know and which has already given you a good return.
There is therefore a risk that the leverage you draw will give you less money than the current optimal leverage you know and which you are certain will give you such a return.
From this experience, we easily understand that information (collected by trying another machine in the casino) is valuable, but it has a price (since the new machine can disappoint us). Experimentation is sometimes a form of “misinvestment” but it is this experimentation that will potentially lead you to the best machine in the casino.
The exploit/explore trade off
Without getting too into mathematics, the solution to the bandit’s problem is easy to understand: the optimal strategy is to start with a period of exploration, where you pull levers at random and collect information.
When you have more information about what works and what doesn’t, you spend most of your time leveraging (exploiting), but you continue to explore other options in case your best current option isn’t the best one that exists.
The singular thing to understand is that the exploration phase virtually never stops. Even if, deep down, you are sure you have found the best possible option, you never stop experimenting, because the information you collect while experimenting is always valuable.
In other words, the only way to beat the bandit is to keep trying new things.
Life is a bandit problem
There are a number of decisions you can make that will help you get what you want in life. You have no guarantee that everything you choose will produce the desired results, and you start with very little information about what is best for you as an individual.
However, you have one major advantage: other people play the same game, and you can look at what they do to gather information about what works and what doesn’t without having to do exactly the same things they have done.
The optimal strategy remains more or less the same: experiment as much as possible, with as many variations as possible, and pay particular attention to the experiences of others. As you find things that seem to be producing the desired results, spend more time and energy doing them. As your efforts produce results and your certainty in this option increases, increase your investment in this option accordingly.
But never stop experimenting: try new options, discover new opportunities, explore new things. The key to a satisfying life is experimentation. The more you experiment, the more you learn, the more information and options you will have at your disposal, and the more likely you will discover the things that will produce the best results for you.
You can’t make positive discoveries that make your life better if you never try anything new.
Start experimenting, and never stop.