It makes a very important reply to anyone who claims that e.g. you should stick with Occam’s original Razor and not try to rephrase it in terms of Solomonoff Induction because SI is more complicated.
Humans and their silly ideas of what’s complicated or not.
What I find ironic is that SI can be converted into a similarly terse commandment. “Shorter computable theories have more weight when calculating the probability of the next observation, using all computable theories which perfectly describe previous observations”—Wikipedia.
It makes a very important reply to anyone who claims that e.g. you should stick with Occam’s original Razor and not try to rephrase it in terms of Solomonoff Induction because SI is more complicated.
I take it you mean in the sense “Really? Look how terrible the original is! You’ve got to be kidding.”
I read this as a reminder not to add anything to that map that won’t help you navigate the territory. How is this not a rationality quote? Are you rejecting it merely because of the third disjunct?
I read this as a reminder not to add anything to that map that won’t help you navigate the territory.
The quote doesn’t say that, this is (only) a fact about your reading.
How is this not a rationality quote? Are you rejecting it merely because of the third disjunct?
I’m not especially impressed with the first two either, nor the claim to be exhaustive (thus excluding other valid evidence). It basically has very little going for it. It is bad epistemic advice. It is one of many quotes which require abandoning most of the content and imagining other content that would actually be valid. I reject it as I reject all such examples.
This is an interesting quote for historical reasons but it is not a rationality quote.
It makes a very important reply to anyone who claims that e.g. you should stick with Occam’s original Razor and not try to rephrase it in terms of Solomonoff Induction because SI is more complicated.
Humans and their silly ideas of what’s complicated or not.
What I find ironic is that SI can be converted into a similarly terse commandment. “Shorter computable theories have more weight when calculating the probability of the next observation, using all computable theories which perfectly describe previous observations”—Wikipedia.
Hrm.… I’m not sure if I can sufficiently state the speed prior in natural language faster than that, without some really good auctioneer training.
I take it you mean in the sense “Really? Look how terrible the original is! You’ve got to be kidding.”
I read this as a reminder not to add anything to that map that won’t help you navigate the territory. How is this not a rationality quote? Are you rejecting it merely because of the third disjunct?
The quote doesn’t say that, this is (only) a fact about your reading.
I’m not especially impressed with the first two either, nor the claim to be exhaustive (thus excluding other valid evidence). It basically has very little going for it. It is bad epistemic advice. It is one of many quotes which require abandoning most of the content and imagining other content that would actually be valid. I reject it as I reject all such examples.