Perhaps ML models should be able to ‘play’ as well.
Bumble bees play, according to new research
51–60 of 79 posts
Re: Bumble bees play, according to new research
#52Paying it forward: An article by David Graeber I first saw here, where he argues the utilitarian evolutionary view (every action witnessed in animals is there to somehow increase changes of survival) is a relic of the industrial revolution. Instead, he says, playing is a key feature of life. https://thebaffler.com/salvos/whats-the-point-if-we-cant-hav...
RIP David
“Debt: The First 5000 Years” is mind boggling and I hold it as high as Sapiens. In some ways it’s actually better when explaining what people actually did instead of our idealized views.
Re: Bumble bees play, according to new research
#53Earlier quoted context omitted.
Some level of intelligence is a prerequisite for experience of qualia. A rock for example is below that threshold. The question then becomes what exactly is the minim level of perception we are willing to accept. Individual cells respond to their environment indirectly, that’s enough to qualify IMO but I accept people would disagree. Which just illiterates people don’t generally use philosophical definitions directly…
The ideas is if you had some, theoretical, mechanical computer built of stone (something trivially producible), if we believe that any artificial intelligences can be sentient, then it would be trivial to extend this to create a sentient, but essentially just a complex stone object.
A single stone is different than a system consisting of large numbers of stones in some non random configuration.
It’s like the flaw in that “Chinese room” thought experiment. The person following instructions isn’t the room any more than the walls are. All that thought experiment demonstrates is a CPU has a more limited perspective than the computer it’s part of.
Re: Bumble bees play, according to new research
#54There used to be a bumble bee that, I was convinced, would come around to play with my dog, a Labrador. It was entertaining to watch. When the dog was outside the bumble bee would hover near his face to get his attention. Once he noticed the bee he would growl, jump, and bite at the bee but not in an aggressive manor. It was in his playful tone, the same as when I played with him and his toys. Every time he lunged at…
Hover flies compete with other makes for the privilege of hovering in the 'best' spot over a bush or wherever. They'll fly at whoever invades their space and then return to hovering and defend it.
Re: Bumble bees play, according to new research
#55Earlier quoted context omitted.
The ideas is if you had some, theoretical, mechanical computer built of stone (something trivially producible), if we believe that any artificial intelligences can be sentient, then it would be trivial to extend this to create a sentient, but essentially just a complex stone object.
A single stone is different than a system consisting of large numbers of stones in some non random configuration. It’s like the flaw in that “Chinese room” thought experiment. The person following instructions isn’t the room any more than the walls are. All that thought experiment demonstrates is a CPU has a more limited perspective than the computer it’s part of.
I mean, showing that difference is extremely hard philosophically.
Re: Bumble bees play, according to new research
#56Earlier quoted context omitted.
A single stone is different than a system consisting of large numbers of stones in some non random configuration. It’s like the flaw in that “Chinese room” thought experiment. The person following instructions isn’t the room any more than the walls are. All that thought experiment demonstrates is a CPU has a more limited perspective than the computer it’s part of.
I mean, showing that difference is extremely hard philosophically.
How so?
A specific rock inherently is whatever that rock is. That same rock can represent all finite messages at the same time in relation to it’s position along arbitrarily defined number lines. Nothing inherent to the rock changed as that meaning is external to it.
Re: Bumble bees play, according to new research
#57Earlier quoted context omitted.
I mean, showing that difference is extremely hard philosophically.
How so? A specific rock inherently is whatever that rock is. That same rock can represent all finite messages at the same time in relation to it’s position along arbitrarily defined number lines. Nothing inherent to the rock changed as that meaning is external to it.
Consciousness is an actual thing. It's not an idea. The idea that somehow a collection of stone objects springs forth a consciousness when arranged in one pattern versus another... is certainly an odd concept.
Re: Bumble bees play, according to new research
#58Earlier quoted context omitted.
How so? A specific rock inherently is whatever that rock is. That same rock can represent all finite messages at the same time in relation to it’s position along arbitrarily defined number lines. Nothing inherent to the rock changed as that meaning is external to it.
Consciousness is an actual thing. It's not an idea. The idea that somehow a collection of stone objects springs forth a consciousness when arranged in one pattern versus another... is certainly an odd concept.
Why would you assume consciousness is a thing rather than a property of a system?
Temperature for example isn’t a substance. I am not saying consciousness couldn’t be a thing, but it just seems to be an assumption without justification.
As to arrangements of stuff being important, arrangement of individual pieces can make the difference between a watch and a pile of pieces or a person and a dead body. Many emergent properties depend on the specific arrangements of components.
Re: Bumble bees play, according to new research
#59Paying it forward: An article by David Graeber I first saw here, where he argues the utilitarian evolutionary view (every action witnessed in animals is there to somehow increase changes of survival) is a relic of the industrial revolution. Instead, he says, playing is a key feature of life. https://thebaffler.com/salvos/whats-the-point-if-we-cant-hav...
> This would be the reaction of most professional ethologists as well.
A behaviorist maybe, but an ethologist?
* an ethologist surely knows that gene mutations can be advantageous for the family/group as well as the individual
* an ethologist surely knows that a previous advantage for survival can persist as a vestige
* an ethologist can surely rinse and repeat to find all kinds of fascinating combinations in living organisms which-- while they may not disadvantage the organism-- cannot be explained in the present as construing any clear survival advantage
Those alone are enough to develop an incremental path out of the narrow rational view that Graeber ascribes to Ethology's history. I can't imagine Ethology didn't already get there.
Ethologists: what say you?
Edit: Related process in an unrelated field-- early 20th Century German comparitive musicology eventually became the perfectly reasonable and relativist field of Ethnomusicology. AFAICT Ethnomusicologists kept the instrument taxonomy and the ethnologies, and then got off the "gold standard" of high-Romantic German art music. :)
Re: Bumble bees play, according to new research
#60There used to be a bumble bee that, I was convinced, would come around to play with my dog, a Labrador. It was entertaining to watch. When the dog was outside the bumble bee would hover near his face to get his attention. Once he noticed the bee he would growl, jump, and bite at the bee but not in an aggressive manor. It was in his playful tone, the same as when I played with him and his toys. Every time he lunged at…
s/manor/manner :)