/explore
|
You are here 
|
reasonableapproximation.net | ||
| | | | |
www.lesswrong.com
|
|
| | | | | One-sentence summary: Kelly is not about optimizing a utility function; in general I recommend you either stop pretending you have one of those, or s... | |
| | | | |
www.greaterwrong.com
|
|
| | | | | There was a twitter thread that triggered some confusion amongst myself and some other people in a group chat I'm in.1 The relevant tweets are these (I've omitted some): 3) Let's say you were offered a coin flip. 75% it comes up heads, 25% it comes up tails; 1:1 payout. How much would you risk? 4) There are a number of ways to approach this question, but to start: what do you want, in the first place? What's your utility function? 5) In other words-how cool would it be to make $10,000? How about $1,000,000-is that 100 times as good? | |
| | | | |
www.lesswrong.com
|
|
| | | | | Suppose you know what outcomes are better than what other outcomes, but not by how much. Given two possible actions you could take,AandB, if you k... | |
| | | | |
www.lesswrong.com
|
|
| | | AI used to be a science. In the old days (back when AI didn't work very well), people were attempting to develop a working theory of cognition. ... | ||
