ProgrammingGodJordan
Banned
Garbage comment.Those items are irrelevant spamming of the thread with non-science github links!.
You may have misses it, but artificial neural networks is a part of modern science.
Super Artificial Intelligence, a naive approach
- Thread starter ProgrammingGodJordan
- Start date
You may have misses it, but artificial neural networks is a part of modern science.
ProgrammingGodJordan
Banned
The above is heavily garbage bound.
(1)
Thought curvature concerns manifolds.
You may have been unaware, but manifolds are potential solvers of severe discrimination issues in modern machine learning.
Each entity in the problem space is observable as detangleable manifolds.
(2)
I didn't express that computer speed was the only thing.
Instead, I mentioned that computer speed was core to the success of modern machine learning.
(3)
Now, the human brain is efficient per general intelligence. (Compared to other brain like constructs)
This efficiency has something to do with how many cycles per frame exist per material.
(3)
10^14 artificial synaptic operations per second are currently done by machines.
As these machines got more powerful, they did more and more cognitive tasks. (Some exceeding human intelligence)
It doesn't take a genius to see that with some number of advancements per power/structure, overall artificial human level intelligence is inevitable.
Last edited:
Reality Check
Penultimate Amazing
Pointing out basic science is never worthless: Dimensional analysis.
Followed by an actually worthless comment from you
. I have read the thread and know where that number comes from.Reality Check
Penultimate Amazing
You have missed the fact that github is not scientific literature
! And that artificial neural networks is not the subject of this thread - it is some undefined ""naïve approach ".Thus: Those items are irrelevant spamming of the thread with non-science github links!
Last edited:
Reality Check
Penultimate Amazing
Calling the real world garbage is not wise. In the real world the OP contains:
A point I am trying to make is that you have not yet described your "naive approach" after a couple of days and 345 posts.
Last edited:
ProgrammingGodJordan
Banned
You have missed the fact that github is not scientific literature
And that artificial neural networks is not the subject of this thread - it is some undefined ""naïve approach ".
Thus: Those items are irrelevant spamming of the thread with non-science github links!
Deepmind is the planet's prominently powerful artificial Intelligence.
Deepmind atari q, is on github.
From that paradigm, advancements were made.
Gihub or not, neural code is science.
ProgrammingGodJordan
Banned
This does not perturb the gihub links' contents to suddenly become non science.
The description of the naive approach is contained in the paper/code, from the original post.
Reality Check
Penultimate Amazing
Manifolds are a wonderful mathematic concept I first came across when learning General Relativity and I do know about their use in machine learning.
ProgrammingGodJordan
Banned
(A)
Based on my original post's source, you can find an estimation of 10^15 synapses
At 10^15 synapses, we have 10 impulses per second, which yields 10^16 sops.
(B)
I had long discussed that Moore's law was decreasing.
Moore's law ends soon after 2020. (Hence the 2020 calculation was pertinent)
Regardless, any rate of improvement of machines shall probably yield artificial human level intelligence.
Reality Check
Penultimate Amazing
A collection of code is not scientific literature, i.e. non-science.
I am not going to register at a web site to download an incoherently titled PDF.
Reality Check
Penultimate Amazing
Gituib is still not scientific literature.
Reality Check
Penultimate Amazing
Read the OP: If mankind isn't erased (via some catastrophe), on the horizon of Moore's Law, mankind will probably create machines, with human-level brain power (and relevantly, human-like efficiency), by at least 2020.
Based on your numbers and Moore's law that claim is invalid. If Moore's law decreases then the claim is even more wrong.
Though you may want to clarify what "on the horizon of Moore's Law" means. It sounds like you expect it the law stop in 2020 and the number of transistors per chip to remains constant forever
.
Last edited:
ProgrammingGodJordan
Banned
That's fine.
...but as I mention, it is common enough in machine learning.
At basis, each x ~ P may be observed as some manifold.
The problem then becomes separation of manifolds, such that unique solutions are discoverable for some dataset/input range, where each matrix representation is a bijective inverse for some following representation, on some continuous function sequence.
In other words, we have a detangling problem.
This is at least the sketch at the manifold level.
Last edited:
ProgrammingGodJordan
Banned
I did the calculations over some range a * 10^15 synapses, on Moore's law 2 years per minimization cycle.
So, the calculations holds while Moore's law is still alive.
Regardless, super intelligence is probably inevitable, given any rate of improvement over speed and or programming/architecture.
Reality Check
Penultimate Amazing
A basic sketch stating what is known does not create super artificial intelligence.
ETA: Even a cartoon does not create super artificial intelligence
.
Last edited:
Reality Check
Penultimate Amazing
The actual calculation shows that it is impossible to achieve your claim by 2020 of roughly 10^16 to 10^18 synaptic operations per second.
The OP links to a 2014 IBM SyNAPSE chip with "only" 256 million synapses.
A claim of 10^14 computer operations a second gets to the lower limit of the range that you gave in a little over 8 years (doubling every 2 years), i.e. at least 2025. But your source for this is not an currently working chip - it is a simulation on a massively parallel supercomputer.
The rest of the post may need a Duh! because it is just about inevitable that we will have "a brain in a box" sometime - maybe within the next few decades.
Last edited:
ProgrammingGodJordan
Banned
I don't see how 10^14 synapses for current machine level, and human 10^15 synapses = 10^16 sops fail to yield 2020 year, applying H = K * 2n.
ProgrammingGodJordan
Banned
A basic sketch stating what is known does not create super artificial intelligence.
ETA: Even a cartoon does not create super artificial intelligence
Here is slightly better description, in loose machine learning terms:
- Points maintain homeomorphisms, such that for any point p under a transition T on some transformation/translation (pertinently continuous, inverse function) t, p0 (p before T) is a bijective inverse for p1 (p after T); on t.
- Following the above, topologies maintain homeomorphisms, for any collection of points W (eg a matrix of weights), under some transition T on some transformation/translation sequence (pertinently continuous, inverse functions) s, W0(W before T) is a bijective inverse for W1(W after T); on s, where for any representation of W, determinants are non-zero.
- Now, topological homeomorphisms maintain, until linear separation/de-tangling, if and only if neural network dimension is sufficient (3 hidden units at minimum, for 2 dimensional W)
Otherwise, after maintaining homeomorphism at some point, while having insufficient dimension, or insufficient neuron firing per data unit, in non-ambient isotopic topologies that satisfy NOTE(ii) W shall eventually yield zero determinant, thus avoiding linear separation/de-tangling. At zero determinant, unique solutions for scalar multiplications dissolve, when the matrix becomes non-continuous, or non-invertible.
NOTE(i): The state of being "ENTANGLED" is the point before which some de-tangleable classes are de-tangled/made linearly separable.
NOTE(ii): Unique solutions in matrices are outcomes that resemble DATA SETS; for homeomorphisms (topologies: where zero-determinant continuous invertible transformations/translations engender OR ambient isotopies: where positive/nonsingular determinants, nueron permutations, and 1 hidden unit minimum occurs, i.e for 1-dimensional manifold, 4 dimensions are required in network)
Last edited:
Reality Check
Penultimate Amazing
I added some text to the posts you replied to but I will emphasize this:
10^14 synapses for current machine level wrong because no current machine has 10^14 synapses.
Even with this imaginary 10^14 synapse machine, you fail to get to 2020.
10^14 * 10 = 10^15 synaptic operations per second.
Double this to get to 2019: 2 * 10^15 synaptic operations per second.
Double this to get to 2021: 4 * 10^15 synaptic operations per second.
That is less than half of the lower limit of your claim. So the claim is debunked even if we ignore the reasonable interpretation of "roughly" + a range as meaning "somewhere in that range"
!Even worse, the mostly likely value in a range of values in the middle of the range, thus:
It is standard in science and common in real life that a calculation on a range of values does not use the extremes or values outside of that range
Last edited:
Reality Check
Penultimate Amazing
Random highlighting does not make a better description of anything. What might be a cut and paste from a textbook is a waste of space - link to the source.
ETA: looks more like a cut an paste from here which hints of out of context, mathematic word salad from an amateur.
Last edited: