RussDill
Philosopher
In the terminology of academic paper, research doesn't mean, "Look stuff up". It means design an experiment, do the experiment, and analyze the results.
Merged Artificial Intelligence Research: Supermathematics and Physics
- Thread starter ProgrammingGodJordan
- Start date
ProgrammingGodJordan
Banned
You words are not entirely true.
ProgrammingGodJordan said:
Apart from a few more things I need to do to complete the experiment proposal as you see in the image above, I can't actually implement the design above; if one reads the thought curvature paper one may quickly find out that the design concerns quantum computation/computers, which I have no access to...
Footnote:
In the world of research, there are countless stretches of time where initial iterations of novel paradigms are first and foremost, actually built (i.e. theories designed by some entity before some separate body performs experiments) prior to being used in the field.
As a great example, this paper by Bengio has no experimental design, let alone experimental results, and yet it may prove quite useful in a future scope.
At least for now, I am still improving the experiment design as seen in thought curvature.
Apart from your inaccurate words, do you have anything to contribute w.r.t. to the OP? (You may use this conversation on thought curvature on science forums as a guide)
Last edited:
abaddon
Penultimate Amazing
Wheee. Copypasta spammage. Just what we need more of. How many times have you barfed that image? I have lost count.
Reality Check
Penultimate Amazing
@ProgrammingGodJordan
I have never stated that I disagree with pursuing AI.
The point I am trying to make is that ignorant, incoherent and obscure PDFs on the Internet are not anyone pursuing AI.
This is a science section of the forum. People here know about science. They know that there are more than 2 people in the world doing AI research
!
Last edited:
ProgrammingGodJordan
Banned
Perhaps if you pursued Ai, the PDF wouldn't appear like gibberish to you.
I've exchanged emails with people like Bengio Yoshua, about the contents of Thought Curvature, and he didn't respond to me as if the paper was gibberish.
At any cost, physics probably won't be complete without describing the observer (i.e. human level intelligence)
Perhaps Tegmark is onto something fruitful...
Reality Check
Penultimate Amazing
A link repeating a delusion that Deepmnd has an "atari q architecture"
15 August 2017: Ignorant nonsense about Deepmind
18 August 2017: Repeated "atari q" gibberish when DeepMind is not an Atari machine and has no "q" (does have Q-learning)
P.S. A small nitpick is that DeepMind is the company. The Atari video games AI is "The Arcade Learning Environment (ALE)" as stated in Playing Atari with Deep Reinforcement Learning. That paper has only 2 DeepMind words (company name and email address).
The latest on ALE: Revisiting the Arcade Learning Environment: Evaluation Protocols and Open Problems for General Agents (September 2017)
5 October 2017: No experiment at the given link or PDF!
The image he has given before that looks like gibberish is not an experiment of any kind. Which leads to:
5 October 2017: A PDF section title lies about a probable experiment.
"A probable experiment: A Transverse Field Ising Spin (Super)–Hamiltonian Quantum Computation" is followed by no experiment at all!
High school science students have a good idea what an experiment is. What an experiment is comprehensively taught to physicists (me in a previous life) in their first year at university. By the time I went onto post-graduate work I had done dozens of experiments so I can recognize one. Since he goes on about QFT - that was a good part of my thesis which was a "Hamiltonian Quantum Computation" on nearly magnetic dilute alloys.
5 October 2017: A link to a PDF repeating a delusion of a "Deepmnd atari q architecture".
15 August 2017: Ignorant nonsense about Deepmind
18 August 2017: Repeated "atari q" gibberish when DeepMind is not an Atari machine and has no "q" (does have Q-learning)
P.S. A small nitpick is that DeepMind is the company. The Atari video games AI is "The Arcade Learning Environment (ALE)" as stated in Playing Atari with Deep Reinforcement Learning. That paper has only 2 DeepMind words (company name and email address).
The latest on ALE: Revisiting the Arcade Learning Environment: Evaluation Protocols and Open Problems for General Agents (September 2017)
5 October 2017: No experiment at the given link or PDF!
The image he has given before that looks like gibberish is not an experiment of any kind. Which leads to:
5 October 2017: A PDF section title lies about a probable experiment.
"A probable experiment: A Transverse Field Ising Spin (Super)–Hamiltonian Quantum Computation" is followed by no experiment at all!
High school science students have a good idea what an experiment is. What an experiment is comprehensively taught to physicists (me in a previous life) in their first year at university. By the time I went onto post-graduate work I had done dozens of experiments so I can recognize one. Since he goes on about QFT - that was a good part of my thesis which was a "Hamiltonian Quantum Computation" on nearly magnetic dilute alloys.
Last edited:
Reality Check
Penultimate Amazing
A lie about an "irrelevant one line description of deep q learning"
That post is you insulting me. The "one line description" is a lie. I did not describe "deep q learning" at all. I cited the Wikipedia DeepMind company article and quoted a partial sentence about their neural network.
You stated that I wrote an "irrelevant one line description of deep q learning". This is a lie because
5 October 2017: A lie about an "irrelevant one line description of deep q learning".
That post is you insulting me. The "one line description" is a lie. I did not describe "deep q learning" at all. I cited the Wikipedia DeepMind company article and quoted a partial sentence about their neural network.
You stated that I wrote an "irrelevant one line description of deep q learning". This is a lie because
- I cited an relevant Wikipedia article.
You mentioned DeepMind so DeepMind is relevant
! - I did not write any description about "deep q learning".
I gave a quote from the Wikipedia DeepMind company article
Last edited:
ProgrammingGodJordan
Banned
No such lie was told.
You ironically just showed that my statement was accurate, where I mentioned that you described deep q learning in a single line.
That partial quoting was a brief line, done by yourself, apparently unbeknownst to yourself.
ProgrammingGodJordan
Banned
Edited by Agatha:
Edited material sent to the split thread
Why is it invalid to express that Deepmind presented a model that could play atari, using q learning, i.e. an atari q architecture?
Reality Check said:
Why say that I expressed an "ignorant statement", where I mentioned that atari q left out pooling layers, when in fact they did leave out pooling layers?
See this breakdown not by me, why atari q left out pooling layers.
Reality Check said:
Reality Check said:
I don't know what you are on about above, but here is a detailed, intuitive, mathematical description of mine, regarding Deepmind's flavour of deep q learning (written in 2016):
https://www.quora.com/Artificial-In...p-Q-networks-DQN-work/answer/Jordan-Bennett-9
Reality Check said:
An experiment is proposed, see a snippet here:
[IMGw=650]https://i.imgur.com/L96rxC3.png[/IMGw]
Beyond designing the above experiment to some degree, I can't do much beyond that, as I don't have access to a quantum computer.
Last edited by a moderator:
Reality Check
Penultimate Amazing
That is still not an experiment. It is not a proposal for an experiment. It is the entire section, not a "snippet".
Reality Check
Penultimate Amazing
I have added to 5 October 2017: A lie about an "irrelevant one line description of deep q learning"
You stated that I wrote an "irrelevant one line description of deep q learning". This is a lie because
- I cited an relevant Wikipedia article.
You mentioned DeepMind so DeepMind is relevant- I did not write any description about "deep q learning".
I gave a quote from the Wikipedia DeepMind company article
ProgrammingGodJordan
Banned
(1) I don't recall mentioning anywhere that it was an actual experiment, but a proposal instead.
I even ensure to note this in the paper; notice the "Probable" term before experiment.
(2) The highlighted portion of your quote above is merely your opinion, fueled by your demonstrably ignorant prior statements; as I said before,
I've exchanged emails with people like Bengio Yoshua, about the contents of Thought Curvature, and he didn't respond to me as if the paper was gibberish.
Notably the nature of Bengio's responses likely hold validity in contrast to yours; you have demonstrated that you weren't aware that deep q models could exclude pooling.
ProgrammingGodJordan
Banned
Also, it is a snippet; the proposal for the experiment concerns the Supermanifold Hypothesis in Deep Learning, which underlines possible structures to be considered in the proposal experiment (i.e. "supersymmetric tensors"), which the image from the thought curvature paper does not include...
Last edited:
Reality Check
Penultimate Amazing
Edited by Agatha:
Edited material sent to the split thread
The Arcade Learning Environment architecture does not any part of the Atari architecture. ALE does not have a "q" architecture. Thus "atari q architecture" is doubly wrong. Anyone who read about the DeepMind company knows this.
DeepMind
ALE has a convolutional neural network, Q-learning with reinforcement learning architecture.
An aside: It would be interesting to see if ALE could be implemented on an Atari machine to play Atari games
!
Last edited by a moderator:
Reality Check
Penultimate Amazing
I read what you wrote. Read what I wrote. I will make it clearer:
5 October 2017: No experiment at all, proposed or actual at the given link or PDF!
5 October 2017: A PDF section title lies about a probable experiment no experiment at all, proposed or actual.
ProgrammingGodJordan
Banned
..and as I mentioned prior, others such as Bengio, don't appear to respond to me in a manner that would underline that the paper is gibberish..
Footnote:
From prior instances, you appear to demonstrate lack of awareness of basic machine learning know how.
For example, on 15 August 2017, when I underlined in the paper, Deepmind's atari q model, that excluded pooling, why did you then go on to discuss some paper that including pooling? (See your original response here)
Your words after the above quote in a later reply: "I already knew about their use in convolutional neural networks so I went looking for their possible use for DeepMind."
Why did you go searching for pooling use in Deepmind models after I had underlined that such convolutional neural net based models excluded pooling in atari q?
Was that not a false intuition of yours that atari q indeed, included pooling, or possibly, that neural nets with convolutional layers could not exclude pooling?
Why go on to search for something that merely included pooling if your expectations were supposedly comprised of models that naturally excluded pooling?
Last edited:
RussDill
Philosopher
Wow, you are much farther out of your league than I had guessed. First it's "I don't have access to GPU compute resources". And now it's "I don't have access to quantum compute resources". How incredibly convenient. Lucky for you, I know a think or two about computation and computability theory.
Mathematicians and computer scientists have an official "measure" of the power of something to be computed. There are various classifications:
https://en.wikipedia.org/wiki/Computability_theory
Anyone who knows enough about quantum theory and Turing machines can quickly come to the realization that a quantum computer is Turing complete, nothing more. In fact, there is a long list of programs that will run your quantum algorithm on a classical computer:
https://www.quantiki.org/wiki/list-qc-simulators
There's even this site on online one with a web interface that uses GPU resources.
The Great Zaganza
Maledictorian
- Joined
- Aug 14, 2016
- Messages
- 29,770
intelligence/cognitive ability and sentience are at best very loosely connected:
single-cell organisms, hives and plenty of machines can react intelligently to changes in the environment - but no one would argue that by making them bigger/faster they would become more sentient.
Complexity also isn't a useful criterion: our brain isn't more complex than a tropical storm, yet no one things of the weather as being sentient.
in short: making A.I. faster and better doesn't mean that it will become more sentient.
single-cell organisms, hives and plenty of machines can react intelligently to changes in the environment - but no one would argue that by making them bigger/faster they would become more sentient.
Complexity also isn't a useful criterion: our brain isn't more complex than a tropical storm, yet no one things of the weather as being sentient.
in short: making A.I. faster and better doesn't mean that it will become more sentient.
Last edited:
ProgrammingGodJordan
Banned
[IMGw=350]https://i.imgur.com/MrxleHs.jpg[/IMGw]
(1) I still lack sufficient GPU resources, in order to do particular tests w.r.t. to certain parts involving thought curvature.
(2) If you read the paper you would find that the paper is based on something called the quantum boltzmann machine, and quantum reinforcement learning.
So the outcome is that I lack yes, both GPU resources, and Quantum Computing resources.
Thanks for trying to help, but you ended up attacking my thread without evidence, like many others here have done.
If you're going to attack, attack with evidence please, and actually take more than 5 minutes to read thought curvature.
Last edited: