[ Removed by moderator ]

0 points by BreadfruitFluid1984 15 hours ago on reddit | 13 comments

All other considerations must be truly equal for Ockham's rule to apply. And it says "usually the simplest explanation is the best" in such cases.

Einstein is quoted (somewhere) "Everything should be made as simple as possible, but no simpler."

So Newton's model of gravitation is simpler than Einstein's. Much simpler. And most of the time it's good enough. (That is accurate enough to accurately describe or explain some phenomena.) But it's not a better explanation than General Relativity.

[OP] BreadfruitFluid1984 | 15 hours ago

So , for example there a child with symptoms of flu, according to Ockham's rule the doctor can simply assume it's just a common flu but the doctor should always check for other possible causes as a caution right I think a video mentioned these types of caution once .?

I don't think Ockham's razor is an excuse for not being thorough.

[OP] BreadfruitFluid1984 | 14 hours ago

Ohhh 😯 ok then thank u for ur answer by the way still new to this field 🐸 Came here cus it's always best to learn from more experienced people u know ? 👌

GlacialFrog | 14 hours ago

It’s a useful rule of thumb, but it isn’t a scientific fact and shouldn’t be treated as one, yet often is by people who don’t know any better.

fox-mcleod | 14 hours ago

A lot of people think it’s a rule of thumb but it isn’t.

Stated formally, it’s possible to prove it mathematically. The formal way to put it is that under common sense assumptions (axioms), the best possible scientific model is the shortest algorithm that generates the empirical data under consideration.

Solomonoff's theory of inductive inference proves it.

Essentially, the proof points out that “simpler” has a rigorous definition: in the sense of Kolmogorov simplicity. And that this definition through information theory can be turned into a probablistic statement.

To get a sense for this, consider a simple special case. The mail shows up in the morning. Which theory is more probably correct:

A) a mail carrier brought it

B) a mail carrier named Gina brought it

Adding excess information beyond what is required to explain the phenomena (the name of the mail carrier is easy to vary and has no explanatory value) always reduces the chances of the statement being true.

You need to close the parenth on the link.

fox-mcleod | 12 hours ago

Thanks!

exclaim_bot | 12 hours ago

>Thanks!

You're welcome!

Sams_Antics | 13 hours ago

You’ll hear it referred to as “all else being equal, the simplest explanation is most likely,” but simplest is misunderstood. It means “makes the fewest and/or least radical assumptions” in this context.

fox-mcleod | 12 hours ago

The best way to understand simplest is “if I were programming a simulation designed to reproduce what we’ve observed, which explanation requires the fewest lines of code?

This is called Kolmogorov complexity and there is actually a mathematical proof that the (successful) theory with the lowest Kolmogorov complexity is the best scientific theory under examination.

ApeOnARockInSpace | 12 hours ago

Historically it’s been notoriously ambiguous. Elliot Sober has a book length treatment of it called Ockham‘s Razors if you want to get into the weeds.