What is the Problem of Numeric Parsimony? (Ockham's Razor)

An explanation of the problem of numeric parsimony, including the distinction between qualitative and quantitative parsimony, and concerns with how quantitative par

Here are some videos you might enjoy:

Philosophy by Topic:

Information for this video gathered from The Stanford Encyclopedia of Philosophy, The Internet Encyclopedia of Philosophy, The Cambridge Dictionary of Philosophy, The Oxford Dictionary of Philosophy, The Oxford Companion to Philosophy, The Routledge Encyclopedia of Philosophy, The Collier-MacMillan Encyclopedia of Philosophy, the Dictionary of Continental Philosophy, and more! (#Parsimony #philosophyofscience)

[loganaragon]

Great timing, seems that a lot of folks are using Ockham’s Razor. Thanks for the video!

[Pfhorrest]

It's neither types nor tokens per se but the detail of your overall description of the world that matters.

If you have one theory that posits that there is exactly one special member of a special class of thing, and another that says there can be indefinitely many things of that type none more special than the other, the latter is more parsimonious because it's less specific; for example, if we have a model of time that specifies that the present is a special unique time and pasts and futures are different than it, that's less parsimonious than if we say that all times are just more instances of the same type of thing as the present time, which is merely *this* one of that general class of thing. Likewise if we have a model of possible and actual worlds that says possible worlds are a different kind of thing than the actual world and there is just one special unique actual world, that is less parsimonious than saying there are just worlds, none more special than any other, and the actual is merely *this* world.

On the other hand, if you need to start specifying that particular instances of objects exist for an explanation, then the more particular instances you have to specify, the more complex your description and so the less parsimonious it is altogether. Positing that there is a single unique planet of a special hard-to-observe kind is unparsimonious; positing that there is a general kind of hard-to-observe planet, one of which could be present in this instance, is parsimonious.

[scottmiller]

Go read up on Minimum Description Length, if you haven't already - it will give (an) answer to all your open questions in the opening. One modification to the usual definition you will see in Wikipedia, etc., is that the MDL needs to include the length of the language used to describe it as well as the length of the object described.

[johncalligeros]

'Seek simplicity then distrust it.' A.N. Whitehead (paraphrased.)

[InventiveHarvest]

With econ training, the question of 'how much' is answered. The answer is always the point where marginal benefit equals marginal cost.

In Ockham's Razor, the benefits, predictive power, are the same and so we are concerned with costs. Presumably, a theory that posits one unobserved planet would cost more to test than a theory that posits two small planets.

Take pi for example. Using the number 3 would be simpler, but wouldn't provide a good estimate. However, calculating 1000 digits of pi would be very costly, but might not provide much benefit over using something like 3.14.

[cubefox]

Nice video. Here is a tip though: The audio quality of the microphone isn't very good. It's not as bad as in old videos but still not exactly pleasant to listen to.

[tomholroyd]

sorites paradox. quantity vs quality, when does that happen