Vlad's Roam Garden

Powered by 🌱Roam Garden

worth of the decision is determined by expected value and not the outcome

People tend to judge the worth of the decision in retrospective [by it's outcome]([[outcome bias]]), but even though good outcome is an evidence of a good decision it does not necessarily stem from good decision. People tend to underestimate prevalence of randomness in life.

The best strategy is the one that has the highest expected value given your current knowledge about the world

There is always uncertainty about how any given decision would play out. This is caused by our lack of knowledge about the world and our inability to predict it. Only Laplace's demon can make a perfect decision.

I find many worlds analysis a powerful way to convey this idea

With the data you have at hand - you should make a decision in such a way that you maximize the proportion of all possible worlds where you achieve your goal.

We don't have control over outcomes but we can make our decisions better

Second, knowing that you are using an optimal algorithm should be a relief even if you don’t get the results you were looking for. The 37% Rule fails 63% of the time. Maintaining your cache with LRU doesn’t guarantee that you will always find what you’re looking for; in fact, neither would clairvoyance. Using the [[Upper Confidence Bound algorithm]] approach to the explore/exploit tradeoff doesn’t mean that you will have no regrets, just that those regrets will accumulate ever more slowly as you go through life. Even the best strategy sometimes yields bad results — which is why computer scientists take care to distinguish between “process” and “outcome.” If you followed the best possible process, then you’ve done all you can, and you shouldn’t blame yourself if things didn’t go your way.
Outcomes make news headlines—indeed, they make the world we live in—so it’s easy to become fixated on them. But processes are what we have control over.