Your ability to judge both yourself, and another person, and how your personalities will interact, is limited. It's sufficient to identify people with whom you absolutely will not get along, with reasonable accuracy; this is low-hanging fruit. So let's say you've eliminated 95% of the candidate pool by this point.

The remaining 5%? You're now considering a pool of candidate partners who you can't immediately eliminate (assuming you have more than one person remaining, after all probably-unsuitable people are eliminated). At this point your list of candidates are people about whom you are uncertain. -Remember- that you're uncertain.

Or, from a different angle: If you are absolutely certain that a relationship with somebody will work out, that sense of certainty should, due to the Dunning-Kruger effect, be taken as evidence that you should be less certain.

The Correct Rational Approach to Finding Life Partners:

Start with two facts: First, the vast majority of women are not, in fact, suitable life partners for you. Second, you are not a suitable life partner for the vast majority of women.

These imply a course of action which starts with elimination. If building an online dating profile? Your goal is not to attract as many suitable people as possible. Your goal is to -reject- as many unsuitable people as possible; this is the entry point for people looking for you, and there are far, far more unsuitable people than suitable people. The same is true in real life, which can be as simple as avoiding locations which are primarily populated by unsuitable people. (Bars, as a rule, for pretty much anybody who would be reading this.)

Likewise, when searching for people, your goal is -rejection-. If you're looking for the hottest girl in the bar - you've already failed, because you're not looking to reject people. Also, you're in a bar. Reject the locations, first. "Is this somewhere I'm likely to meet somebody who fits my interests, who who would be interested in me?" Maximize the ratio of acceptable to unacceptable people.

This is Less Wrong - go to Lindy Hop or otherwise swing dance classes. It's the nerdiest dance community you'll find, and the gender proportions, depending on where you are, will probably favor you if you're male. Also, it will help with your proprioception, which, given that you're on Less Wrong, could probably use some help anyways.

Once you've eliminated the unsuitable, do -not- pick the "best". You're probably pretty good at identifying what won't work, but you're probably pretty terrible at identifying what will.

So be open to short-term flings. These can turn into long-term relationships - although you shouldn't expect them to.

Hell, be open to casual sex. These encounters can -also- turn into long-term relationships - although, again, you shouldn't expect them to.

Be open to friendships. Once again, expect nothing.

In general - once you've eliminated the unsuitable, be open. You're looking for pearls; once you've sorted them out, don't toss the oysters overboard before you've checked. They may surprise you.

Don't seduce people into long-term relationships in any terms, let long-term relationships happen on their own. If it takes a special effort to make somebody fall in love with you, it will take a special effort, constantly, forever, for them to stay in love with you.

Information diet?

I did a quick search on LW but didn't find any important article about information diet. Did I miss something?

Questions worth considering:

• Should we eliminate all news sources like some advocate?
• What about the news that are relevant, e.g. changes in the tax code that you need to know about?

So I'm aiming for the soft spot of eliminating all the unnecessary news while still getting those pieces that are relevant for me.

Any ideas?

A measurement is an observation that quantitatively reduces uncertainty.

A measurement reduces expected uncertainty. Some particular measurement results increase uncertainty. E.g. you start out by assigning 90% probability that a binary variable landed heads and then you see evidence with a likelihood ratio of 1:9 favoring tails, sending your posterior to 50-50. However the expectation of the entropy of your probability distribution after seeing the evidence, is always evaluated to be lower than its current value in advance of seeing the evidence.

So then this initial probability estimate, 0.5, is not repeat not a "prior".

This really confuses me. Considering the Universe in your example, which consists only of the urn with the balls, wouldn't one of the prior hypotheses(e.g. case 2) be a prior and have all the necessary information to compute the lookup table?

In other words aren't the three following equivalent in the urn-with-balls universe?

1. Hypothesis 2 + bayesian updating
2. Python program 2
3. The lookup table generated from program 2 + Procedure for calculating conditional probability(e.g. if you want to know the probability that the third ball is red, given that the first two balls drawn were white.)

Good books on economics, investing?

Are there equivalent books to "Probability theory, the logic of science" and/or "The Feynman lectures on Physics" in economics or investing?

Who are the great authors of these fields?

Not quite -- there not a lot of empirical data and the biases and incentives are much more... attenuated in AI than in investing :-)

The usual :-)

• Look at empirical data
• Don't believe everything you're told
• Adjust for known biases and incentives

etc. etc.

This seems a bit more appropriate for the open thread. Or searching; I've argued frequently that you should Just Buy An Index Fund, and things like this article explain how you might be able to do a bit better than that.

Seconding what Vaniver said, and I'm pretty sure the efficient market hypothesis applies as much to forums dispensing investment advice as it does to investing itself. And at least if you were looking for individual gurus, there wouldn't be the possibility of Eternal September turning the advice quality to dirt once the forum becomes known.