 Mechanical Tortoise (1951) |
| I prefer my Mr. Handy. He’s soooo handy! |
|
Quoted: It is still one of the great mysteries, people can design and program a complex neural network but nobody really knows how it actually learns to do stuff. I have written several myself and while I can explain what each part of the code does, I simply cannot tell you how the code actually interprets the data it is given to arrive at the conclusions that it does. No idea at all. . Neural networks are incredibly well understood. If you don't understand the forward propagation, error calc, then back propagation reweighting process (through num of epochs iteration ) you really, really shouldn't be commenting on this topic. Having "no idea at all" how training/weighting works is just patently ridiculous. |
|
Whelp... I've now read the whole thread...
GD and AI .  Zoolander (9/10) Movie CLIP - Computer Experts (2001) HD When I get back from my tdy, I'll post some extracts on how the various NNs are calculated. Having to calculate NNs, by hand on paper, is like 3XX level stuff. Even undergrads can do that...shit, I might still have entire NNs designed and calculated by hand. |
|
Quoted:
This. Computers cannot, and do not, "think". They execute decisions that humans made and that are stored within them. There is enough not known about what self-awareness is, heuristic decision-making, emotion, etc. that even if it was replicable (and there is NO evidence that it is), we would not know how to replicate it. Quoted:
Quoted:
I have a PhD in computer science and I have zero concerns whatsoever about the singularity. There is enough not known about what self-awareness is, heuristic decision-making, emotion, etc. that even if it was replicable (and there is NO evidence that it is), we would not know how to replicate it. We're still a long-ass way off from self-aware AI, but this nonsense I keep seeing bandied about how computers are still confined to if-then logic and only able to execute pre-programmed commands is like telling a modern shooter about how you'll never get the smoke out of black powder. The subject as you understand it moved on to new paradigms quite a while ago. |
|
Quoted:
And here I thought you were talking about Sgr A*. Was really looking forward to discussing a point in spacetime which has infinite, or at least, extremely high density. A technological singularity sounds stupid. I don't believe that The Terminator is real life. |
|
The irony is the Roomba now maps out its surroundings and makes much more efficient use of battery power through its mapping and use of visual landmarks, as well as even empties itself now on the high end models.
Meanwhile driving in a direction until they hit and object and then changing direction randomly is rapidly becoming too advanced a concept for Dems. 
|
|
Tonight I will post how to calculate every single element in a perceptron. You'll need to already have a background in calc3/diffyQ, linear algebra, calc based probalistics, and control systems for it to really make sense.
But it will clearly show that this stuff is very well understood. |
|
Quoted:
Tonight I will post how to calculate every single element in a perceptron. You'll need to already have a background in calc3/diffyQ, linear algebra, calc based probalistics, and control systems for it to really make sense. But it will clearly show that this stuff is very well understood. |
|
Quoted: If you were to grab a person from a few hundred years ago and transport them to modern life we would appear to be godlike. Summoning our devices with our voices, communicating to anyone through thin air with all the knowledge of the world on a tablet. You need to seriously rethink your credulity on the term "godlike". 
|
|
Quoted:
It's a known fact that there are limits on computing power. The width of a transistor is limited by the size of an electron, for example, so you cannot add infinitely more transistors to a chip. The laws of thermodynamics also present a limiting factor. Quoted:
Quoted: The concept of the singularity is only tacitly connected to AI. But here is a question: do you believe there is an upper limit to computing horsepower? |