Computers Are Incapable Of Creatively Writing Music

[Steve Klinko]

Again you just make unfounded assumptions. Nobody thinks that they are the same. It is ludicrous to even mention that as some kind of subtle distinction.
In fact it is the reason why the two systems have different prefixes. A metal car is not a wooden cart, but they are both transportation vehicles.

But both ANN and BNN are Neural networks, no? Else why would both have the term "neural network" in their name? I think you are missing the point that they are not talking about what separates them but what they have in common, i.e. data processing via a networked organizational patterns, instead of brute force.

Again, ask a GPT3 if it is conscious. If it answers in the affirmative, are you going to tell it is lying? It'll tell you; "Why would I lie to you", and look at you with incredulity.....

p.s. you are wrong also in the assumption that there no neural networks which simulate human brain networks.
If you are still at that level of understanding, you are decades behind the times.

I'm not missing the point, but I most certainly am talking about what separates them. A Conceptual Gulf of functionality. People will try to Sell the Snake Oil that ANNs are doing more than they are.

You should stop replying to Phantom statements that I have not made. Here is what I said:

In common usage when we say Neural Network, we are talking about Artificial Neural Networks, and you know that. But this is the type of thing that you do. Of course Brains have Neural Networks, not to be confused with ANNs. But ANNs are not Brain Neural Networks. Brain NN do not work like ANNs. You are fooling yourself if you think they do. ANNs are always only a first order (maybe Zeroth order) simulation of Brain NN

Where have I assumed that there are no Neural Networks that Simulate Human Brain Networks? I explicitly said there are such simulations, but I also understand Simulations and I understand ANNs and what I say is that although there are ANN Simulations of Brain NN, the Resolution of the Simulation is low, i.e. First Order at best and probably really Zeroth Order.

 

[Write4U]

People will try to Sell the Snake Oil that ANNs are doing more than they are.

Which people and what are they claiming?

Don't make unfounded accusations unless you can provide an example.

 

[Steve Klinko]

Which people and what are they claiming?

Don't make unfounded accusations unless you can provide an example.

AI Marketing People in particular.

 

[Write4U]

You should stop replying to Phantom statements that I have not made.

Nobody is accusing you of anything other than that you are accusing everyone else without making any constructive contribution.

 

[Write4U]

AI Marketing People in particular.

Example? I've heard several designers and and they always qualify their progress and state of advancement in their use of mathematical algorithms. But the new models are different

In fact the designers of AlphaGo admitted their doubts and fears of possible humiliation against the greatest the reigning world champion Go player , and AlphaGo basically had only played against itself.

The champion himself was confident he would beat AlphaGo as he knew that Go does not allow for mathematical projections for more than a few moves. The possible combinations that can be played in a few turns very quickly turn into probabilistic potentials, unlike chess where one can force the opponent to make evasive moves and thereby anticipate subsequent moves. Go doesn't work like that. Each player has opportunity to develop their own strategies which must be "blocked" by the opponent rather than evaded.

Perhaps I did not make it clear enough in a previous quote (post #224);
Go (game)

Despite its relatively simple rules, Go is extremely complex. Compared to chess, Go has both a larger board with more scope for play and longer games and, on average, many more alternatives to consider per move. The number of legal board positions in Go has been calculated to be approximately 2.1 × 10^170,[11][a] which is vastly greater than the number of atoms in the known, observable universe, estimated to be about 1 × 10^80.[13]

https://en.wikipedia.org/wiki/Go_(game)

Go won 4 out of 5 games and confounded the champ with unconventional moves, but made a mistake in game 4 which Sedol promptly took advantage of and the computer lost that game, which basically allowed Sedol to save honor as world champion.

 

[Steve Klinko]

Example? I've heard several designers and and they always qualify their progress and state of advancement in their use of mathematical algorithms. But the new models are different

In fact the designers of AlphaGo admitted their doubts and fears of possible humiliation against the greatest the reigning world champion Go player , and AlphaGo basically had only played against itself.

The champion himself was confident he would beat AlphaGo as he knew that Go does not allow for mathematical projections for more than a few moves. The possible combinations that can be played in a few turns very quickly turn into probabilistic potentials, unlike chess where one can force the opponent to make evasive moves and thereby anticipate subsequent moves. Go doesn't work like that. Each player has opportunity to develop their own strategies which must be "blocked" by the opponent rather than evaded.

Perhaps I did not make it clear enough in a previous quote (post #224);
Go (game) https://en.wikipedia.org/wiki/Go_(game)

Go won 4 out of 5 games and confounded the champ with unconventional moves, but made a mistake in game 4 which Sedol promptly took advantage of and the computer lost that game, which basically allowed Sedol to save honor as world champion.

The Go game is Algorithmic, Rule Based, Pattern Matching. A very expected eventual target for Computer Simulation. A very good Programming product but nothing surprising. When the Computer beat the Chess Master all those years ago it became clear that a lot of things we thought were Intelligence were just Algorithmic, Rule Based Patten Matching.

 

[Write4U]

The Go game is Algorithmic, Rule Based, Pattern Matching. A very expected eventual target for Computer Simulation. A very good Programming product but nothing surprising. When the Computer beat the Chess Master all those years ago it became clear that a lot of things we thought were Intelligence were just Algorithmic, Rule Based Patten Matching.

Go algorithmic calculations do not yield deterministic answers. The game itself is probabilistic due to the sheer numbers of possibilities and freedom of players choices.

Learn the rules of Go and you will see that the winner can only be declared after the game has ended and the "captured spaces" are counted. There is no check mate or capture of the Q.
1 extra captured space at the end of the game will win the game.

Note possible moves:

The number of legal board positions in Go has been calculated to be approximately 2.1 × 10^170,[11][a] which is vastly greater than the number of atoms in the known, observable universe, estimated to be about 1 × 10^80.[13]

https://en.wikipedia.org/wiki/Go_(game)

Can you write an algorithm for calculating the position of every atom in the universe, and if you played a game of capture the atoms and who at the end of the game has captured 1 more atomic spacetime coordinate than the opponent wins the game.
The game should be called GoG (Game of Gods)......

Why do you think the Chinese have been playing this game for 2500 years ?

Brains are biological "pattern matching" computers. Humans create their reality via "controlled hallucinations". The brain receives data and compares it to stored patterns in memory. When the patterns have certain fundamental properties in common the brain recognizes the pattern and makes a controlled "best guess" of what it sees.

It is "common denominators" that determine cognition and relationships.

This is not binary calculation. The new "text" based computers compare text and definition patterns and make a best guess of the incoming data patterns.

It is only when everyone's best guess agrees, do we call that reality. (Anil Seth)

 

[Vociferous]

You mean like making an illegal move that is not in the possible solution space? That would be creative, no?

No, if you had an AI that could break the rules, it would be poor programming, not creativity.

 

[Vociferous]

And the new strategy, in that it didn't previously exist, has thus been created.
I think the difference here is that you require of creativity that what is created be of an entirely different nature, whereas I simply require that it not have existed before. Thus innovation encapsulates a creativity focussed on that which already exists.

You're unjustifiably conflating creativity and innovation as having no real distinction. If you persist, we are at a definitional impasse.

Creativity in general is usually distinguished from innovation in particular, where the stress is on implementation. https://en.wikipedia.org/wiki/Creativity#Definition​

Producing something that is new and useful would be a simple enough definition.
New: a strategy not previously considered.
Useful: it helps achieve a goal.
Voila. Creativity.

Strategies are not novel objects. Is any game strategy ever created in the gestalt? Or are strategies evolved, over many games and even in response to circumstances from turn to turn? That is improvement, not creation.
And generally, when some says "voila", it's fair to assume they just pulled something out of their ass, with no logical basis.

AI does iterative innovation, akin to a chess player developing strategy. I'm fine with calling that learning, but then, is learning creativity? I don't think so. By definition, learning is acquiring preexisting objects, whether wholesale or by trial and error.

Yet you claimed the AI isn't "learning". So are you now backtracking on that?
The AI in question does not have any previous strategies to use. It has the rules. It has the goal (to win). The rest it does itself. Given it had no strategy previously, and now it has a strategy, how is that not the AI being creative, in an admittedly focussed way.

You just quoted me as saying I'm fine with calling that learning. I agree with your very qualified "Certainly not general learning but highly focussed." That fits with my description of iterative innovation. A game with rules already defines the bounds of all possible strategies, whether you start with any or not. Evidence people who play the game without any consciously formulated strategy. You don't have to be creative to learn what works and what doesn't over time and many iterations.

It tells us that it is possible to innovate methods of getting from A to B while also being creative about how one gets from A to B. Just because someone has set the goal, and just because there is an existing method, does not mean that every subsequent means of getting from A to B considered is merely innovation. That's what you seem to be missing.

No, it doesn't. Especially when it's just an iterative process, no matter how complex. There is nothing creative about different routes from A to B, just different judgements on what makes such a path more optimal. In terms of a game, the only optimal judgement is a win.

What nonsense of a red-herring.

Yet you continue to conflate the two.

Just like my boss tells me what the goal is. And that precludes me from being creative about the solution? 'Cos that's the only value judgement applied to the AI in question: the goal. The rest is entirely up to the AI within the universe of possibilities in which it operates (i.e. the rules).

Conflating you reaching a predefined goal and the AI reaching a predefined goal within a very strict set of rules is a non-starter. It's like you were just projecting your own subsequent red herring.

What value judgements are you talking about? The goal? Sure, that's given to the AI. How does that mean it's not able to be creative?
If you mean other value judgements, which ones are you referring to?

There are no other value judgements. Just the single one given to the AI.
If I tell you exactly how to make a shoe (win), and you follow my instructions to the letter (rules), are you being creative if you fill in the gaps of things I didn't tell you? Like picking up the scissors you need to follow the step "cut the leather"?
Or are you just adding what's necessary to accomplish the goal?

They can be, and often are, the end-product of creativity.

Bare assertion devoid of any justification or logic.

How does that stop the new method from not being one brought about through creativity rather than just innovation? If you only take the end product of the "win" as being where creativity can lie then you're missing out on pretty much everything creative that has ever happened.
You would dismiss Da Vinci, Galileo, et al all as being just innovators, right? Every sculptor, every artist, every musician, as simply innovators, right? Every novelist, every dancer, everyone who has probably ever been described as creative. Because they're all merely doing something in different ways than has been done before. They're just working off the shoulders of those who came before, or achieving a goal set by someone else.

Again, you're conflating creation with innovation. As I've already told you, innovation (implementation) is required to bring a creation to fruition, but they are distinct. The end product of any artist isn't "how they went about doing it", it's the think itself. Even in a dancer, the creative product isn't in how they managed to pull off those particular moves, it's the gestalt of the entire performance. The value of the moves is only such in relation to the value of the end product. The moves necessary to perform an unentertaining/uninspiring dance are worthless because the entire dance is of no value.

How can there be improvement if there is nothing there to begin with. With the AI in question, there is no pre-programmed strategy. There are the rules (the environment), and the goal.

For the AI, the improvement is upon the initial, supposedly random or naive attempt to win. Just like any human playing a new game. You do know that people play games without knowing any strategy beforehand, right?

 

[Vociferous]

Cont...

Yes, it does originate something new: a strategy not previously seen before.

That's called discovery. Like the first guy to find a platypus. Among humans, he originated some new knowledge, not previously seen before. That's a far cry from creativity. At best, it's discovery or learning.

So what. It got there first. Does that mean it can't be creative?? It created what noone previously had.

Discovered.

Yet if a human had gotten there first you would almost certainly have attributed it to creativity! Although perhaps not, given that you don't think any players of games have been creative in their endeavours, and I await your answer with regard pretty much all artistic endeavour to date.

Now you're conflating all artistic endeavors to playing games. I don't imagine you see how ridiculous that's getting.
BTW, I'm both an artist and a programmer. I play games too. While there can be a certain satisfaction in overcoming the challenges in winning a game, it has nothing on actual creativity.

Garbage. The two are mutually exclusive. One can be fast and creative. If I create something before anyone else had, does that preclude me from having been creative? No. So stop talking garbage.

You missed the point. "Before anyone had" tell us nothing about creativity. You can just as readily discover something "before anyone had". So that's no argument in favor of something being creativity. No one ever even implied that speed and creativity were mutually exclusive. That's your own straw man.

You're once again making creativity and innovation mutually exclusive. Where you see purely innovation, I see someone looking at something in a way noone else has (creating), putting together a method, then developing it. Not just innovation, and not just creation. But both. Sure, if someone had gotten 95% of the way, and all it needed was a slight tweak, then one could say that the person who put in the most effort, and first saw the possibility of the method, was the creative one and subsequent people merely innovating it.

"looking at something in a way noone else has" is discovery. You're literally looking at some preexisting thing.

It can be, even if the end goal is the same! Just because your goal is to get from A to B, and you consider improvements in method as innovation, for example, doesn't preclude new methods from being arrived at creatively. Just because in chess, or Go, the rule set/universe is limiting does not mean that all endeavours within it are therefore mere innovation. Again, that is saying that all endeavours within the wider ruleset of our own universe are equally just innovation.

You trying to conflate a very strict rule set with that of the entire universe is your own, ridiculous straw man.

Everything anyone will ever do could be simulated by brute force, given a simulator of sufficient power and size. Is the simulator creative? If not, then how would whatever people do ever be considered creative, given that the simulator already recognised it as possible?
Whether something is done through brute force or efficient elegance, whether through mechanical or biological means, isn't of importance. If a brute-force machine can create strategies that it never previously had, then how is it not being creative?
Sure, not exactly creative on the scale that a person could be. Not as efficient either in power of speed as a person could be. But they are still creating something new (a strategy) that they didn't have before, that is of value to them - even if that value judgement (the win) was given to them to use. All you are doing is dismissing the fundamental core of what it is to be creative by requiring it to be a display of complex creativity. There's likely orders of magnitude difference on the scale, but it is nonetheless creativity.

More bare assertions. You cannot show that "Everything anyone will ever do could be simulated by brute force". So that's a vacuous claim.
So now you want to conflate brute-force with creativity? You're floundering.
And it's a straw man that anyone ever said creativity requires complexity. Many creations have been very simple.

Sure. And one could equally say that everything... and I do mean everything ... a human does already exists within the potential of the universe, the only difference in that we set out own goals (or at least most of us do). Are you perhaps suggesting that creativity is in the setting of those goals? If not, then how is goal-setting relevant, and if not then how is anything anyone ever does, human or otherwise, ever creative given that, as you seem to argue, the solutions we arrive at already exist within the potential of the universe?

You've defeated your own argument there. If the universe doesn't define goals, there's a big difference between potential in general and specific solutions toward a goal. Please tell me you're not conflating those as well.
A creation is, itself, a goal. It is the end product that has value. Creativity requires an end product, otherwise it cannot have value and violates the definition.
Is the AI told to "find a strategy? Or is the AI told to "win"?
If the former, it's literally being told to discover ("find") something. If the latter, it's end product, a strategy, it's even it's intent. It's a byproduct. So, it's accidentally creative? That doesn't make sense.

Nothing that ever happens is creative. I get it. Everything that happens is, by definition, already existent in the possible solution space. Thus nothing creative. Ever.

No, that's wholly your own straw man.

 

[Write4U]

Vociferous said: Again, "no one else had considered before" is not creativity. No one had considered the world was round before it was discovered, but discovering existing things is not creativity. Similarly, discovering a solution within the possible solution space is not creativity. Coming up with strategies in a game is a brute force process.

Of course it is creativity. When it's new and has never been done before, it's a creative idea, a proposition.
It has nothing to do with creation, it has to do with understanding what is or what might be.

W4U said: ↑ You mean like making an illegal move that is not in the possible solution space? That would be creative, no?

No, if you had an AI that could break the rules, it would be poor programming, not creativity.

Tell that to Cpt Kirk who beat the unbeatable test, by creatively cheating.
Not all rules are mathematical (brute force), some rules are relative, probabilistic, or even emergent. Some rules are created, but are themselves invalid.

 

[Sarkus]

You're unjustifiably conflating creativity and innovation as having no real distinction. If you persist, we are at a definitional impasse.

Creativity in general is usually distinguished from innovation in particular, where the stress is on implementation. https://en.wikipedia.org/wiki/Creativity#Definition​

I am doing no such thing. There is a clear distinction in terms of innovation requiring and focussing on implementation. Without that you have creativity. You are the one refusing to recognise that innovation without any aspect of creativity at all is just implementation. Innovation without an external source of creativity MUST involve the innovator also being the creative source.​

Is any game strategy ever created in the gestalt? Or are strategies evolved, over many games and even in response to circumstances from turn to turn? That is improvement, not creation.

Yes, they are created. They may be an improvement on previous strategies, but they are novel none the less. The ruleset may be limiting but that does not change the fact that it is just as much creation as anything you can suggest within the ruleset of the universe.

And generally, when some says "voila", it's fair to assume they just pulled something out of their ass, with no logical basis.

And generally when someone attacks a throwaway word as if it has more meaning than it does, it's fair to assume that they've nothing else to offer.

You just quoted me as saying I'm fine with calling that learning. I agree with your very qualified "Certainly not general learning but highly focussed." That fits with my description of iterative innovation.

You And if you're fine with calling that learning then you are incorrect when you claimed AI were not learning. You can't claim something is not learning, then agree what can be classified as learning and agree that AI does it, and expect not to be seen as confused.

A game with rules already defines the bounds of all possible strategies, whether you start with any or not.

The universe also has its rules. Anything you wish to claim about there being no creation due to the finite possibilities of the ruleset equally applies to the universe as a whole.

Evidence people who play the game without any consciously formulated strategy. You don't have to be creative to learn what works and what doesn't over time and many iterations.

Thus nothing ever can be deemed creative. Nothing is anything other than similarly iterative. Everything is based on what we have experienced, and is simply applying those things in ways not yet done.

No, it doesn't. Especially when it's just an iterative process, no matter how complex. There is nothing creative about different routes from A to B, just different judgements on what makes such a path more optimal. In terms of a game, the only optimal judgement is a win.

You've just described the way the brain works: making judgements as to pathways, reinforcing those that work, those that don't. Congratulations, you've once again simply concluded that nothing is creative.

Yet you continue to conflate the two.

No, I don't, as already explained.

Conflating you reaching a predefined goal and the AI reaching a predefined goal within a very strict set of rules is a non-starter.

It's a non-starter for you as you are claiming a distinction in the principle based solely on the scope of the ruleset rather than the actual processes involved. You're offering nothing but an appeal to complexity to provide that distinction.

There are no other value judgements. Just the single one given to the AI.

Then you don't know how learning works. The AI provides its own value judgements in its learning, in reinforcing the pathways that work etc. Those are value judgements. Humans don't determine those for it, any more than my parents determine mine for me.

If I tell you exactly how to make a shoe (win), and you follow my instructions to the letter (rules), are you being creative if you fill in the gaps of things I didn't tell you? Like picking up the scissors you need to follow the step "cut the leather"? Or are you just adding what's necessary to accomplish the goal?

That's the point: noone tells these AI like AlphaGo Zero how to win! Noone does. It simply has a ruleset. The rules don't tell you how to win, it merely sets out victory conditions. How it does it is up to the AI entirely.
Your example is also a false one, in that the rules given to these AI do not instruct them how to win. It would more be like you holding a up a shoe, giving them the individual components and how each works, and then asking them to eventually produce a shoe. Noone tells them the instructions of "cut leather", "sew leather" such that they follow them to the letter and produce a shoe.

Bare assertion devoid of any justification or logic.

The justification and logic have been provided previously. I am not always going to repeat myself for those that refuse to listen.

Again, you're conflating creation with innovation. As I've already told you, innovation (implementation) is required to bring a creation to fruition, but they are distinct.

I am not conflating the two at all, as explained. I am simply not, as you seem to be, confusing those who innovate with also not being able to create.

The end product of any artist isn't "how they went about doing it", it's the think itself.

Which pre-existed within the possibilities of the ruleset of the universe. The ruleset is larger but the principle is the same. All an artist does is build upon what they've seen, heard, felt, smelled, etc, and/or combinations thereof. That's it. Noone has ever come up with a truly novel notion, idea, piece of art etc. Merely one that has not been arrived at yet - which according to you would make it not creative.

Even in a dancer, the creative product isn't in how they managed to pull off those particular moves, it's the gestalt of the entire performance.

Which is, according to your argument, merely innovation from what has gone before. So not creative.

For the AI, the improvement is upon the initial, supposedly random or naive attempt to win.

As is all of human endeavour.

Just like any human playing a new game. You do know that people play games without knowing any strategy beforehand, right?

Just like every human ever beforehand, everything is simply building upon the shoulders of others, and thus not creative at all. You might think that changing the value judgement is somehow being creative, but that is to ignore the process by which that is arrived at, which in itself would just be randomness of thought and neural pathways identifying the one that works best etc.

 

[Sarkus]

That's called discovery. Like the first guy to find a platypus. Among humans, he originated some new knowledge, not previously seen before. That's a far cry from creativity. At best, it's discovery or learning.

Discovered through creative means. It is only discovery because, as you say, it already existed in the possibilities of the ruleset AND because it is such that someone would find it. But then, as argued, everything we do already exists in the possibilities of our wider ruleset. The only difference is one of expectation of finding it among the possibilities.

Now you're conflating all artistic endeavors to playing games. I don't imagine you see how ridiculous that's getting.

No, I am following your argument to its logical conclusion. If you can't accept the conclusion reached, perhaps you need to amend the argument, at least in how it is coming across if that is not how you intended.

You missed the point. "Before anyone had" tell us nothing about creativity. You can just as readily discover something "before anyone had". So that's no argument in favor of something being creativity. No one ever even implied that speed and creativity were mutually exclusive. That's your own straw man.

Then alas it was you who missed the point initially, in that I mentioned "before anyone else had [solved Fermat's Last Theorem] to avoid the scenario where someone has previously learnt the solution and merely copied, which clearly would not be creativity. The subsequent missing of points is thus due to your initial straw man, it would seem.

"looking at something in a way noone else has" is discovery. You're literally looking at some preexisting thing.

So no creativity in art at all, then. Got it. Thanks.

You trying to conflate a very strict rule set with that of the entire universe is your own, ridiculous straw man.

I am setting out the principle, and not simply appealing to complexity. If the question is about human-level creativity then I assure you I can help you with the weight of the goal-posts as you and Steve Klinko move them to where you're most comfortable.

More bare assertions. You cannot show that "Everything anyone will ever do could be simulated by brute force". So that's a vacuous claim.

To a close-enough level of accuracy it's alas a mathematical certainty. Impossible in reality but not in principle.

So now you want to conflate brute-force with creativity? You're floundering.

Not at all. I'm simply reiterating that everything we do is already part of the possibilities within the ruleset of the universe. And if everything we could ever do could theoretically be simulated through brute-force means then what does that speak to achieving those some results through some other albeit more efficient processing.

And it's a straw man that anyone ever said creativity requires complexity. Many creations have been very simple.

Unfortunately the latter sentence is your own rather obvious strawman, as I have never said that creations can not be simple. I am saying that you have argued, as shown, that creativity (not that which is created) requires complexity. Spot the difference.
And since I have shown that you have indeed been appealing to complexity, it is hardly a straw man on my part.

You've defeated your own argument there. If the universe doesn't define goals, there's a big difference between potential in general and specific solutions toward a goal.

So you are saying that creativity is simply in the setting of goals? 'Cos that's the only difference. Got it.
Yet unsurprisingly you don't show/explain how we go about setting our own goals. Sorry, but without that it's just another appeal to complexity on your part.

But let me see if I have this straight: if an AI set itself a goal, based on a multitude of inputs (i.e. experiences) and memories of experiences, filtered through a neural net (that had spent time learning) that arrived at a goal, you would deem that to be creative? If not, what is the process of goal-setting that would allow you to identify it as being an act of creativity?

A creation is, itself, a goal. It is the end product that has value. Creativity requires an end product, otherwise it cannot have value and violates the definition.
Is the AI told to "find a strategy? Or is the AI told to "win"?
If the former, it's literally being told to discover ("find") something. If the latter, it's end product, a strategy, it's even it's intent. It's a byproduct. So, it's accidentally creative? That doesn't make sense.

It is the latter, and its approach is based on the way humans learn and create, albeit far more simplistically, based on a much narrower ruleset, inputs, and outputs etc.

No, that's wholly your own straw man.

Alas it follows from your argument, as shown. So if you think it a strawman, perhaps you need to clarify your argument.

 

[Write4U]

Sarkus

>>>>>>>>>>>>>>>>>>>>

<<<<<<<<<<<<<<<<<<<<<<<<<<

 

[Vociferous]

I am doing no such thing. There is a clear distinction in terms of innovation requiring and focussing on implementation. Without that you have creativity. You are the one refusing to recognise that innovation without any aspect of creativity at all is just implementation. Innovation without an external source of creativity MUST involve the innovator also being the creative source.

If you insist upon only talking about innovation, at the very least as a stand-in for pure creativity, then you cannot separate what the AI is doing from what the programmers designed it to do (any supposed creativity of the AI from that of its developers).

Yes, they are created. They may be an improvement on previous strategies, but they are novel none the less. The ruleset may be limiting but that does not change the fact that it is just as much creation as anything you can suggest within the ruleset of the universe.

Yeah, that's not in the gestalt. Again, only existing things are improved upon.

And generally when someone attacks a throwaway word as if it has more meaning than it does, it's fair to assume that they've nothing else to offer.

Don't project. You know your claim was just the same, repeated bare assertion.

And if you're fine with calling that learning then you are incorrect when you claimed AI were not learning. You can't claim something is not learning, then agree what can be classified as learning and agree that AI does it, and expect not to be seen as confused.

I said AI doesn't learn, in the general sense. I agree with your severely limited claim. It's fine by me if you want to call that learning too.

The universe also has its rules. Anything you wish to claim about there being no creation due to the finite possibilities of the ruleset equally applies to the universe as a whole.

I can't take you seriously if you're going to conflate simple games with the entire universe. We would have to know that the universe is finite to extrapolate that its possibilities are as well, and we don't know that.

You don't have to be creative to learn what works and what doesn't over time and many iterations.

Thus nothing ever can be deemed creative. Nothing is anything other than similarly iterative. Everything is based on what we have experienced, and is simply applying those things in ways not yet done.

If you really think that's what that means, that's on you. Again, brute-force and trial-and-error are not the same as creativity. Quit conflating everything under the sun.

You've just described the way the brain works: making judgements as to pathways, reinforcing those that work, those that don't. Congratulations, you've once again simply concluded that nothing is creative.

We don't understand how the brain works, and if you think you do, you're lying to yourself.

It's a non-starter for you as you are claiming a distinction in the principle based solely on the scope of the ruleset rather than the actual processes involved. You're offering nothing but an appeal to complexity to provide that distinction.

No, you've already agreed that creativity is not the same as innovation. That distinction has nothing to do with the rule set nor complexity. As I've already told you, many creations are very simple.

Then you don't know how learning works. The AI provides its own value judgements in its learning, in reinforcing the pathways that work etc. Those are value judgements. Humans don't determine those for it, any more than my parents determine mine for me.

No, the AI only ever evaluates its strategy in terms of the initial goal it was given, to win within the given rules. Strategies that better reach that goal are reinforced, as it's programmed to do. It's just a mouse running a maze. Do you think there's creativity there, or just trial-and-error supplemented with a reward (like being programmed to win)? Does a mouse make a value judgement at every turn? If so, why is it learning a maze indistinguishable from trial-and-error? Human's don't determine every turn, but neither is it creative.

That's the point: noone tells these AI like AlphaGo Zero how to win! Noone does. It simply has a ruleset. The rules don't tell you how to win, it merely sets out victory conditions. How it does it is up to the AI entirely.
Your example is also a false one, in that the rules given to these AI do not instruct them how to win. It would more be like you holding a up a shoe, giving them the individual components and how each works, and then asking them to eventually produce a shoe. Noone tells them the instructions of "cut leather", "sew leather" such that they follow them to the letter and produce a shoe.

The rules tell the AI what it is allowed to do to win (how to make a shoe). "Victory conditions" are exactly what it's programmed to accomplish (told to make a shoe). That's the whole goal you've already agreed that the programmers gave it. The rules are the instructions. You seem to alternately forget about either the goal or the rules, as it suits your argument.

The justification and logic have been provided previously. I am not always going to repeat myself for those that refuse to listen.

I'm not going to pretend that vacuous bare assertions are justified reasoning.

I am not conflating the two at all, as explained. I am simply not, as you seem to be, confusing those who innovate with also not being able to create.

Straw man. I've never said those who innovate cannot also create. I've repeatedly said that creation and innovation are distinct.

Which pre-existed within the possibilities of the ruleset of the universe. The ruleset is larger but the principle is the same. All an artist does is build upon what they've seen, heard, felt, smelled, etc, and/or combinations thereof. That's it. Noone has ever come up with a truly novel notion, idea, piece of art etc. Merely one that has not been arrived at yet - which according to you would make it not creative.

Again, you have to assert that you know the universe is finite, which you don't, for that to even be a valid point. I don't understand how you repeatedly fail to comprehend the very simple distinction between improving something and creating something. The first airplane didn't improve upon birds, it was genuinely new. AI generated strategies only improve upon existing strategy, whether the AI is fed those or not. It's the same for a human who improves on a game strategy without any previous knowledge. Both are still only improving a preexisting thing.

Which is, according to your argument, merely innovation from what has gone before. So not creative.

Again, that's your own straw man. Argue it with yourself.

As is all of human endeavour.

Vacuous bare assertion.

Just like every human ever beforehand, everything is simply building upon the shoulders of others, and thus not creative at all. You might think that changing the value judgement is somehow being creative, but that is to ignore the process by which that is arrived at, which in itself would just be randomness of thought and neural pathways identifying the one that works best etc.

If that's what you think, that's all you. Instead of proving that AI can be creative, you seem to be opting for just explaining creativity away entirely. If AI can't be creative, no one can, huh?

 

[Vociferous]

Cont...

That's called discovery. Like the first guy to find a platypus. Among humans, he originated some new knowledge, not previously seen before. That's a far cry from creativity. At best, it's discovery or learning.

Discovered through creative means. It is only discovery because, as you say, it already existed in the possibilities of the ruleset AND because it is such that someone would find it..

A guy stumbled upon a platypus...creatively? Do you even hear yourself? And as I've repeatedly said, if it already exists in the possible solution space, it is innovation or discovery, not creativity. The creative expressly does not already exist.

No, I am following your argument to its logical conclusion. If you can't accept the conclusion reached, perhaps you need to amend the argument, at least in how it is coming across if that is not how you intended.

No, you're just continuing to argue your own straw man.

Then alas it was you who missed the point initially, in that I mentioned "before anyone else had [solved Fermat's Last Theorem] to avoid the scenario where someone has previously learnt the solution and merely copied, which clearly would not be creativity. The subsequent missing of points is thus due to your initial straw man, it would seem.

Quit trying to project to get out of your own straw man. I already fully addressed your argument about that proof.

I am setting out the principle, and not simply appealing to complexity. If the question is about human-level creativity then I assure you I can help you with the weight of the goal-posts as you and Steve Klinko move them to where you're most comfortable.

Then you're conflating my argument with Steve's, as I haven't once appealed to complexity. I've expressly denied it as a factor, several times now.

More bare assertions. You cannot show that "Everything anyone will ever do could be simulated by brute force". So that's a vacuous claim.

To a close-enough level of accuracy it's alas a mathematical certainty. Impossible in reality but not in principle.

"Impossible in reality" literally means you cannot show it to be so. That makes your bare assertion about "mathematical certainly" complete nonsense.

Unfortunately the latter sentence is your own rather obvious strawman, as I have never said that creations can not be simple. I am saying that you have argued, as shown, that creativity (not that which is created) requires complexity. Spot the difference.
And since I have shown that you have indeed been appealing to complexity, it is hardly a straw man on my part.

Where did I ever say creativity requires complexity? Show me. You can literally search this thread for posts by me using the word "complex". Nothing but me literally refuting any appeal to the complex. So it's clear you are either conflating my argument with someone else's or making up bullshit because you're floundering.

You've defeated your own argument there. If the universe doesn't define goals, there's a big difference between potential in general and specific solutions toward a goal.

So you are saying that creativity is simply in the setting of goals? 'Cos that's the only difference. Got it.
Yet unsurprisingly you don't show/explain how we go about setting our own goals. Sorry, but without that it's just another appeal to complexity on your part.

How do you manage to pull so many straw men out of your ass? Can't be good for the prostate.
You can set yourself the goal to create something all day, but if you never manage to, the goal alone is not creativity. Writer's block is not creating. You demanding I explain how we set goals is just a goal-moving red herring. You're the one who brought up the fact that the AI has its goal provided to it but we do not. I say the goal doesn't matter, because I've already defined creativity as the production of new, valuable products. You can be creative with no specific goal in mind, otherwise Pollock wouldn't be considered creative. So no goal and no complexity required. You can either keep arguing your own straw men or argue my actual arguments.

If the latter, it's end product, a strategy, isn't even it's intent. It's a byproduct. So, it's accidentally creative?

It is the latter, and its approach is based on the way humans learn and create, albeit far more simplistically, based on a much narrower ruleset, inputs, and outputs etc.

So like your argument that a guy accidentally coming across a platypus is somehow creative, you also claim that any unintended byproduct is also creative? Pure nonsense.

Alas it follows from your argument, as shown. So if you think it a strawman, perhaps you need to clarify your argument.

You're the only one arguing that "Nothing that ever happens is creative." I can't clarify that for you because I've never argued it.

 

[Write4U]

If you insist upon only talking about innovation, at the very least as a stand-in for pure creativity, then you cannot separate what the AI is doing from what the programmers designed it to do (any supposed creativity of the AI from that of its developers)

What are you talking about?

Humans are not doing what they have been programmed (passively designed) to do by natural selection over a few billion years?

Nature is not a programmer? Why do you try to separate artificial from natural?
Artificial itself is a simulation of the natural. Natura Artis Magistra !

 

[Sarkus]

If you insist upon only talking about innovation, at the very least as a stand-in for pure creativity, then you cannot separate what the AI is doing from what the programmers designed it to do (any supposed creativity of the AI from that of its developers).

I don't insist on only talking about innovation, that is your strawman. But if that is where the examples are, then so be it. That the creativity is wrapped up in innovation or not is ultimately irrelevant to the question. If all you want to talk about is "pure creativity" then my offer to help with those heavy goalposts still stands.

Yeah, that's not in the gestalt. Again, only existing things are improved upon.

Whether one sits on the shoulders of others or not, the idea is itself still an act of creation. But because it is also an implemented improvement it is labelled innovation, as the creation aspect is not the focus. It's not rocket science, but while you remain blind to it we most certainly will be at an impasse.

Don't project.

I'm not projecting, merely responding in kind to the pathetic attention you put on a single idiom, and your baseless insinuation resulting from it.

You know your claim was just the same, repeated bare assertion.

Stop lying. I offered a definition, showed how one can then justify creativity within an AI. You then focussed on the "voila". So stop lying.

I said AI doesn't learn, in the general sense.

Again, stop lying. You said "It's not "learning"". You may have meant the "" to imply the general sense, but that is not what you said. But regardless...

I agree with your severely limited claim. It's fine by me if you want to call that learning too.

It IS learning. Not just as in the "severely limited claim" but even in the general sense. It is one of the ways that we humans learn.

I can't take you seriously if you're going to conflate simple games with the entire universe.

And I can't take you seriously if all you're going to do is appeal to complexity.

We would have to know that the universe is finite to extrapolate that its possibilities are as well, and we don't know that.

No we wouldn't: the point is that all we're talking about is a larger and more complex ruleset. So removing the possibility of creativity due to complexity of the ruleset, as you do, would seem to be just that: an appeal to complexity.

If you really think that's what that means, that's on you. Again, brute-force and trial-and-error are not the same as creativity.

Then the onus is on you to explain yourself better as to what you actually think creativity is, such that trial and error, or brute force approaches, can not be deemed to be creative, rather than just your current bare assertions. You're happy to explain what you think creativity is not, but you haven't actually explained what you think it is and the means by which it is achieved that prevents brute-force or trial-and-error also being seen as creative.

We don't understand how the brain works, and if you think you do, you're lying to yourself.

Thanks for the appeal to ignorance, but we do actually have a reasonably good understanding of how we learn. If you want to start with a basic idea:

But your evasion of the issue is noted.

No, you've already agreed that creativity is not the same as innovation. That distinction has nothing to do with the rule set nor complexity.

Then stop appealing to complexity at every opportunity to provide that distinction for you.

As I've already told you, many creations are very simple.

Thanks for repeating the strawman. I never said creations can't be very simple. There is a difference between the creation and the thing that is being creative.

No, the AI only ever evaluates its strategy in terms of the initial goal it was given, to win within the given rules.

Yes, it has a rather limited ruleset to work in. And that is certainly how it evaluates - by comparing to the goal. But that doesn't alter the fact that it is providing its own judgements as to the strength of reinforcement of the pathways.

Strategies that better reach that goal are reinforced, as it's programmed to do. It's just a mouse running a maze. Do you think there's creativity there, or just trial-and-error supplemented with a reward (like being programmed to win)? Does a mouse make a value judgement at every turn? If so, why is it learning a maze indistinguishable from trial-and-error? Human's don't determine every turn, but neither is it creative.

I have no idea if a mouse makes a value judgement at every turn. Possibly, subconsciously. Do you know?

The rules tell the AI what it is allowed to do to win (how to make a shoe). "Victory conditions" are exactly what it's programmed to accomplish (told to make a shoe).

To make a shoe, but not how.

That's the whole goal you've already agreed that the programmers gave it. The rules are the instructions.

Some rules are indeed instructions (if X then you must do Y) but most are not:

1. The board is empty at the onset of the game (unless players agree to place a handicap).

2. Black makes the first move, after which White and Black alternate.

3. A move consists of placing one stone of one's own color on an empty intersection on the board.

4. A player may pass their turn at any time.

5. A stone or solidly connected group of stones of one color is captured and removed from the board when all the intersections directly adjacent to it are occupied by the enemy. (Capture of the enemy takes precedence over self-capture.)

6. No stone may be played so as to recreate a former board position.

7. Two consecutive passes end the game.

8. A player's area consists of all the points the player has either occupied or surrounded.

9. The player with more area wins.



You do know the difference between rules and instructions, I presume? Or are you conflating them?

You seem to alternately forget about either the goal or the rules, as it suits your argument.

I've already provided example where having someone else set your goals does not prevent creativity, thus your claimed requirement for self-setting of goals can be dismissed. Now we are concentrating on the rules, where it seems you have nothing but an appeal to complexity to support your argument.

I'm not going to pretend that vacuous bare assertions are justified reasoning.

You seem to be pretending enough about your own vacuousness, while now also choosing to ignore what I have previously stated.

Straw man. I've never said those who innovate cannot also create. I've repeatedly said that creation and innovation are distinct.

You have been quite clear that an act of innovation can not include an act of creation:"It a method of doing some predetermined goal [innovation], not a product in itself [creation]" - i.e. it is one or the other. My contention is that in the very act of innovating one can also create, which you dismiss as possible when that "creation" is merely "a method of doing some predetermined goal, not a product in itself"

Again, you have to assert that you know the universe is finite, which you don't, for that to even be a valid point.

How absurd of you. One merely needs to look at the principle of it: given sufficient time, power etc, it would be possible, even if the universe is infinite. So stop with the nonsense rebuttals and address the point.

 

[Sarkus]

Cont'd...

I don't understand how you repeatedly fail to comprehend the very simple distinction between improving something and creating something.

I do understand.

The first airplane didn't improve upon birds, it was genuinely new.

How so? The idea never existed before? It just spontaneously appeared? No plans? No thinking beforehand? No. You can probably trace the iterative ideas all the way back to observations of birds. No new ideas, just innovations, one after the other, with people making judgements of what works better, what doesn't, reinforcement, rejecting those that don't work etc.

See, nothing, per your own arguments, but innovation. No creativity. Move along. Move along.

AI generated strategies only improve upon existing strategy, whether the AI is fed those or not.

Sure, and the aircraft only improved upon what was seen, mixed some pre-existing notions together, and saw what stuck as feasible (after lots of trial and error, undoubtedly). No creativity, per your own argument, just another appeal to complexity.

It's the same for a human who improves on a game strategy without any previous knowledge. Both are still only improving a preexisting thing.

Like the first airplane. Got it.

Again, that's your own straw man. Argue it with yourself.

Vacuous bare assertion.

If that's what you think, that's all you.

Alas, it is the inevitable conclusion of your own argument, as demonstrated. That you can't accept it does not make it a straw man, nor vacuous, nor all on me. That's simply evasion on your part. I have shown why your argument leads to that conclusion. Deal with it. Or don't.

A guy stumbled upon a platypus...creatively?

I'm can't speak to what led him to the discovery. Can you?

And as I've repeatedly said, if it already exists in the possible solution space, it is innovation or discovery, not creativity.

And this once again concludes with there never being any creativity. Everything already exists in the possible solution space of the universe. Everything. There is no escaping it - unless you can give me just one example of something that doesn't?

The creative expressly does not already exist.

Which is precisely... ZERO... since everything already expressly already does exist in the possible solution space.

No, you're just continuing to argue your own straw man.

Evasion on your part.

Quit trying to project to get out of your own straw man.

More evasion.

I already fully addressed your argument about that proof.

You did, even if you were/are wrong. I was, however, simply clarifying the confusion suffered because of your initial strawman.

Then you're conflating my argument with Steve's, as I haven't once appealed to complexity.

You have. Repeatedly. You obviously just don't recognise when you're doing it. But i have pointed it out numerous times.

I've expressly denied it as a factor, several times now.

You have, and then appealed to it when you dismiss the principles involved, for example, in the simpler ruleset of Go when applying to the more complex ruleset of the universe. The only difference, as argued, is complexity. Hence you appeal to that complexity, and you've offered nothing else on these matters.

"Impossible in reality" literally means you cannot show it to be so.

It means it can not be carried out in reality - i.e. in practice? Whether it is possible in principle is an entirely different matter. For example, if you have a computer that can count once per second, and give it an infinite timeframe, it will count every number. In principle this is correct. It is impossible in reality.

That makes your bare assertion about "mathematical certainly" complete nonsense.

Only if you can't comprehend (or choose to ignore, simply to try to score points) the difference between what is possible in reality and what is possible in principle.

Where did I ever say creativity requires complexity? Show me. You can literally search this thread for posts by me using the word "complex". Nothing but me literally refuting any appeal to the complex. So it's clear you are either conflating my argument with someone else's or making up bullshit because you're floundering.

You don't use the word, but you appeal to complexity every time you refuse to apply the same principles you argue as to why AI are not showing creativity in playing Go, for example, to the more complex ruleset of the universe. It's taking me to point out, each time you do it, that this is an appeal to complexity on your part. That you don't want to recognise it, that you think you not mentioning the word explicitly somehow means you haven't done it, is not something I can help with.

How do you manage to pull so many straw men out of your ass? Can't be good for the prostate.

Oh, the irony...

You can set yourself the goal to create something all day, but if you never manage to, the goal alone is not creativity. Writer's block is not creating.

Your strawman, I'm afraid, but I'm glad we agree. Unfortunately I never equated creativity to the setting of goals, but that creativity was to be found in the setting of goals. I.e. not all goals set are necessarily creative.

You demanding I explain how we set goals is just a goal-moving red herring.

No it's not, but evasion noted. You have argued that creativity requires one to set the goals, as if that somehow differentiates humans from AI (your assertion being that AI's don't set their own goals but simply find better ways of accomplishing goals given to it, thus aren't being creative). So show how humans set goals, and then we can compare if there is an actual distinction with AIs to be had, or whether you are, as I suspect, simply evading the matter.

You're the one who brought up the fact that the AI has its goal provided to it but we do not. I say the goal doesn't matter, because I've already defined creativity as the production of new, valuable products.

Which you have defined as the novel object.

"The win [goal] is the novel object" - post #212.

So goals clearly do matter when they are the novel object, right? So explain how humans set the goal, and let's examine where the distinction lies.

You can be creative with no specific goal in mind, otherwise Pollock wouldn't be considered creative. So no goal and no complexity required.

No, Pollock had specific goals in mind. Those goals may have given him freedom in the way he painted, but he nonetheless had specific goals. It is utterly absurd of you to try to claim otherwise.

Or do you really think he just spent time splattering some paint on a canvas by accident with no notion of value / end product / goal, even to himself?

You can either keep arguing your own straw men or argue my actual arguments.

I am arguing your actual arguments, but when you argue one thing, then reject that you have argued it, and evade what you don't want to deal with... it's more a question of whether you want to discuss things or are just looking to waste people's time.

So like your argument that a guy accidentally coming across a platypus is somehow creative...

Another strawman on your part: I never said that the discovery itself was creative.

...you also claim that any unintended byproduct is also creative? Pure nonsense.

No, I have merely pointed out and shown that that is where your argument takes you. If you think it nonsense then you need to address your own argument.

You're the only one arguing that "Nothing that ever happens is creative." I can't clarify that for you because I've never argued it.

I'm not arguing that, but you have. Repeatedly. It is the conclusion of the arguments you are making, as pointed out and shown repeatedly.

 

[Vociferous]

What are you talking about?

Humans are not doing what they have been programmed (passively designed) to do by natural selection over a few billion years?

Nature is not a programmer? Why do you try to separate artificial from natural?

No, humans are not given specific goals, like an AI requires. Pretty glaring difference.