The Hard Problem of Gravity

[Steven Howard]

I just have issue with this part. That is exactly what the neural network did.

Assume we use periods, commas, quotes, spaces, and the letters A through Z in upper case, to produce a joke that is 8000 characters long or less. If the joke ends early, right pad with spaces. An extremely gracious ceiling on said jokes would be 30^8000. There's no difference between creating a joke in this set and selecting a joke from this set.

It is not possible that the neural network would have produced a completely meaningless string of letters and punctuation marks.

 

[yy2bggggs]

You don't see the difference between creating an algoritihm for jokes and rolling a 30^8000 sided die?

Yes, I do. I also see a difference between selecting one of 30^8000 options and rolling a 30^8000 sided die.

It is not possible that the neural network would have produced a completely meaningless string of letters and punctuation marks.

Is there a point here besides self flattery?

 

[RandFan]

Yes, I do. I also see a difference between selecting one of 30^8000 options and rolling a 30^8000 sided die.

The algorithm is working toward a goal, to string together words and phrases that will likely elicit an emotional response.

Perhaps I'm missing something.

 

[RandFan]

It is not possible that the neural network would have produced a completely meaningless string of letters and punctuation marks.

Yes, unless the algorithm is working toward a goal of producing words and phrases that will likely elicit a response based on certain rules and parameters.

Unless I'm missing something, people forget exactly how modern computers and algorithms solve problems.

 

[cyborg]

Yes, I do. I also see a difference between selecting one of 30^8000 options and rolling a 30^8000 sided die.

The difference is?

 

[yy2bggggs]

The algorithm is working toward a goal, to string together words and phrases that will likely elicit an emotional response.

Perhaps I'm missing something.

Simply that selection doesn't entail random selection or the brute force approach that westprog mentioned.

Here's what westprog wrote:

As I pointed out earlier, as jokes can be encoded in digital form, and they are of finite length, all possible jokes exist in potentia, and it's a matter of selecting them.

There's nothing wrong with what he said. Every joke under the constraints I mentioned, no matter how creative you are at coming up with it, is a member of that set. Being really creative in coming up with a member of that set simply constitutes an effective algorithm for finding really good jokes from that set.
ETA: Sorry, meant to say "there's nothing wrong with what he said here"; leaving original in this edit.

This:

A computer would find it very simple (if tediously time-consuming) to print out every possible joke. Placing them in order of humourousness would be another matter.

...is the thing that I believe you wanted to address, and I agree--I can't see the relevance of the existence of a brute force approach.

 

[RandFan]

Simply that selection doesn't entail random selection or the brute force approach that westprog mentioned.

Here's what westprog wrote:

There's nothing wrong with what he said. Every joke under the constraints I mentioned, no matter how creative you are at coming up with it, is a member of that set. Being really creative in coming up with a member of that set simply constitutes an effective algorithm for finding really good jokes from that set.

This:

...is the thing that I believe you wanted to address, and I agree--I can't see the relevance of the existence of a brute force approach.

A.) It's not simply brute force.
B.) Humans would not likely come to any consensus as to the order of humor.

Again, I think people fail to understand how modern computers and algorithms work. It's not simply brute force and I'm not sure why you think it is. Programmers have been working out problem solving that isn't simply brute force for close to two decades now if not more (that's just how long I've been aware of it).

Here, let me give you an example of what I mean. Early code breaking algorithms used only brute force to find the correct key. Using statistics we could accurately predict how long it would take for a given algorithm to routinely break codes by calculating the number of combinations based on the length of the key. There was nothing at all controversial about this and it was well understood. This held for years until an algorithm was devised that wasn't based on simple brute force and the time to crack the code was significantly reduced. The only way the algorithm could reduce the time needed to find the key is if it did more than simply rely on brute force.

Human cognition is largely problem solving. To be sure we don't typically solve most problems the way computers typically have done so. But programmers aren't stupid. Some time ago they started to find faster and more efficient ways to solve problems using algorithms. I'll let you in on a secret. Google isn't simply a brute force algorithm.

I don't know what your level of understanding is and I don't mean to patronize. Do you understand my point and is it relevant to your's?

 

[yy2bggggs]

The difference is?

The literal difference is that you don't need a die to select. The figurative difference is that not all selection (e.g., natural selection--and yes, that's bona fide selection) is random.

ETA:

Note that rolling a 30^8000 sided die is an algorithm.

 

[yy2bggggs]

A.) It's not simply brute force.

What westprog was suggesting is. I also don't think he was saying that this was the only way to come up with "every possible joke". But I question its relevance.

And a computer can, very simply, brute force.

westprog:

I could have sworn I mentioned this earlier, but just in case...

We are computers.

 

[RandFan]

What westprog was suggesting is. I also don't think he was saying that this was the only way to come up with "every possible joke". But I question its relevance.

And a computer can, very simply, brute force.

westprog:

I could have sworn I mentioned this earlier, but just in case...

We are computers.

Yeah, I suspected that I was simply not getting your point. I appologize.

I can be thick but I honestly try.

 

[BabyHeadedMan]

Funny, I go the opposite direction. I don't think the brain is a magic box, so I think we're eventually going to be able to explain every little bit of what we call "cognition" (which, as I said, is a fuzzy combination of two very different sets of definitions, one of which is decidedly dualistic) in terms of behavior. In fact, we've done a tremendous job in that direction already.

I'm sorry, but I don't really understand what you mean. "Cognition" simply refers to information processing in the brain.

"Experience" as used here is not a term I use at all other than to show it is a metaphorical noun that reifies the behavior of experiencing; "cognition" is a term I use a little as possible because of the fuzziness, preferring "private behavior" as encompassing the portions of cognition that can actually be observed rather than merely inferred from outcomes. If I seem to conflate the two in my writing, it is perhaps from dismissing both.

This is exactly the error made by people of your point of view upon which I'm trying to elucidate. By saying our description of experience itself is merely a reification of the consequences of a certain cognitive process, you are allowing the concept of experience to enter into your model of the mind. Your defining the concept of experience as an instance of reification has no explanatory value in understanding what things we can actually observe about the mind, namely behaviors and the cognitive processes which trigger them.

 

[RandFan]

I'm sorry, but I don't really understand what you mean. "Cognition" simply refers to information processing in the brain.

This is exactly the error made by people of your point of view upon which I'm trying to elucidate. By saying our description of experience itself is merely a reification of the consequences of a certain cognitive process, you are allowing the concept of experience to enter into your model of the mind. Your defining the concept of experience as an instance of reification has no explanatory value in understanding what things we can actually observe about the mind, namely behaviors and the cognitive processes which trigger them.

I don't see an error but then I don't think Mercutio is reifying anything. What we think of as abstract is simply the end result of brain functions much like flight is the end result of wings, feathers and airpressure.

 

[BabyHeadedMan]

I don't see an error but then I don't think Mercutio is reifying anything. What we think of as abstract is simply the end result of brain functions much like flight is the end result of wings, feathers and airpressure.

My model of the mind is that certain cognitive processes cause certain behaviors, but I say nothing about experience itself because assertions of that nature would contribute nothing to what we can observe experimentally, so I consider the nature of experience to be unknowable. Physicalism makes the link between cognition and behavior as well but then goes on to extraneously assert that our description of experience is equivalent to the cognitive processes associated with behaviors which imply experience, such as someone verbally expressing their seeing the color red. The idea of a sensation of red and the idea that there is a pattern of neurons activated by a certain signal from the optic nerve that causes people to do certain things are different from one another descriptively. With what can we show these two descriptions to be referencing the same thing when the answer to the question of the relationship between the two descriptions would contribute nothing to what can be observed experimentally? If cognition and experience are the same thing, we're observing a cognitive process causing a behavior. If cognition and experience are two different things, we're still observing a cognitive process causing a behavior. This is why nothing, in my opinion, can be said about the nature of experience.

 

[yy2bggggs]

This is why nothing, in my opinion, can be said about the nature of experience.

This is wrong. Here are some things I can say about what it's like for me to see red:

• It's closer to what it's like to see orange than what it's like to see blue.
• It's more like what it's like to see blue than what it's like to hear middle C
• It seems impossible to see red without extent and location

There are things we can say about experience. Just dissect them--break them up, and analyze them, and you'll see that experience is comprised of categorizations with aspects under which they are similar to other categorizations and aspects under which they differ.

In other words, experience is, in itself, a form of reasoning.

Now, I don't propose to have a full theory of experiences--but I think they are worthy targets of study, and that there's a lot of very useful information we can get as a result of such study.

 

[Mercutio]

I'm sorry, but I don't really understand what you mean. "Cognition" simply refers to information processing in the brain.

So, you don't understand my point of view. ("Cognition", as I said earlier, has traditionally had at least two very different definitions. One of them, I have no real problem with, although it isn't particularly helpful; the other is out-and-out misleading.)

This is exactly the error made by people of your point of view upon which I'm trying to elucidate.

So, it is a function of my point of view. Which you admit you don't understand.

By saying our description of experience itself is merely a reification of the consequences of a certain cognitive process, you are allowing the concept of experience to enter into your model of the mind.

If you do not know what my model of the mind is, how can you say this?

Your defining the concept of experience as an instance of reification has no explanatory value in understanding what things we can actually observe about the mind, namely behaviors and the cognitive processes which trigger them.

Cognitive processes are one of two things--fictions or behaviors. Private behaviors are not the cause of public behaviors; rather, both private behaviors (the reasonable part of "cognitive processes") and public behaviors are triggered by environmental stimuli.

 

[Mercutio]

My model of the mind is that certain cognitive processes cause certain behaviors, but I say nothing about experience itself because assertions of that nature would contribute nothing to what we can observe experimentally, so I consider the nature of experience to be unknowable. Physicalism makes the link between cognition and behavior as well but then goes on to extraneously assert that our description of experience is equivalent to the cognitive processes associated with behaviors which imply experience, such as someone verbally expressing their seeing the color red. The idea of a sensation of red and the idea that there is a pattern of neurons activated by a certain signal from the optic nerve that causes people to do certain things are different from one another descriptively. With what can we show these two descriptions to be referencing the same thing when the answer to the question of the relationship between the two descriptions would contribute nothing to what can be observed experimentally? If cognition and experience are the same thing, we're observing a cognitive process causing a behavior. If cognition and experience are two different things, we're still observing a cognitive process causing a behavior. This is why nothing, in my opinion, can be said about the nature of experience.

How, precisely, are we observing a cognitive process causing a behavior? At best, the cognitive process is part of the mechanism of the behavior, not the cause. How, not why.

 

[RandFan]

My model of the mind is that certain cognitive processes cause certain behaviors, but I say nothing about experience itself because assertions of that nature would contribute nothing to what we can observe experimentally, so I consider the nature of experience to be unknowable.

Do you have a parsimonious explanation of how the private experience of person A would be any different from the private experience of person B? I suppose you can evoke, again, your claim that is unknowable and therefore it is as likely to be the same as different or non-existent at all.

So, in the end, I don't know if you even exist. I can't disprove idealism and I can't prove that you have private experiences. For that matter I can't "know" that you have such. I guess that leaves solipsism as the only viable philosophy for you, I mean me, or, if I don't exist and you do, then you. And further, you are simply reading these words which are really just emanating from somewhere in your psyche.

Assuming you are real let me ask you, do you find your particular world view fun and satisfying? Should others adopt it? Oh wait, never mind...

What's your point again?

ETA: If you don't get my point then let me state plainly that your philosophy leads to either absurdity or a rejection of everything that seems real.

 

[rocketdodger]

I just have issue with this part. That is exactly what the neural network did.

Assume we use periods, commas, quotes, spaces, and the letters A through Z in upper case, to produce a joke that is 8000 characters long or less. If the joke ends early, right pad with spaces. An extremely gracious ceiling on said jokes would be 30^8000. There's no difference between creating a joke in this set and selecting a joke from this set.

That depends how "selection" and "creation" is defined.

I think you will find that the vast majority of humans consider "selection" the process of choosing from among instantiated entities whereas "creation" is the actual instantiation of an entity to begin with.

As such, "selection" would only occur after "creation," and one would only select from among the already "created" instances.

EDIT: This is probably why so many intelligent forum members spoke up when you said this -- they thought you meant something other than what you did.

 

[rocketdodger]

In other words, experience is, in itself, a form of reasoning.

I am glad we are on the same page now.

 

[rocketdodger]

This:

...is the thing that I believe you wanted to address, and I agree--I can't see the relevance of the existence of a brute force approach.

That's because your worldview doesn't insist you cling to the delusion that all non-human information processing is done by brute force algorithms.