Friday, July 17, 2015
Saturday, July 4, 2015
Saturday, June 13, 2015
Monday, May 18, 2015
Thursday, April 23, 2015
Saturday, February 28, 2015
Friday, January 31, 2014
Tuesday, December 31, 2013
Sunday, May 2, 2010
We can now also consider the dilation described above to be a valid choice of f, if we let g(x)=1/8. The total mapping doesn't preserve density as we required in our simple model of Newtonian mechanics, but now we have a mapping which can be used to describe a gas whose container has expanded in volume by a factor of 2 in all dimensions. (This transformation is still forbidden if we set g(x)=1, which corresponds to the matter-creating transformation which was forbidden in the structured-Newtonian regime.)
It's worth noting that the rotation and translation maps from the Newtonian regime cannot have a decrease in density everywhere in the matter-preservation model: that is, we can't have g(x)<1>
Saturday, December 6, 2008
So we all know that solving the Halting problem is impossible (i.e. it's not decidable). It's also impossible to create a program that is capable of analyzing other computer programs because this problem is equivalent to the Halting problem. (i.e. The program being analyzed could have an infinite loop and there is no way to detect this infinite loop because of the Halting problem).
If we cannot create a program that can analyze other programs then we cannot create a machine that is capable of learning from other machines. In order for a machine to learn from the behavior of other machines it has to be programmed in such a way so that it could analyze other programs and we already said that this is impossible.
One fundamental property of our mind is its ability to learn by, for example, analyzing the behavior of other human beings.
If we cannot create a machine that is capable of learning then would not that mean that there is little hope for duplicating our minds through technology?
Programming a neural net is not a big hairy deal. The easiest way to get a good grounding is to log onto Cornell's open source courseware and look for something in about 4th year computer science.
1) A basic understanding of probability and markov chains
2) The ability to write a tiny little module that takes a single input value and returns one of up to three or four values with stated probability. This will be your node, and you will have as many of them as you can put up with in your net.
3) The ability to use object oriented programming (to create a node class that replicates them a bit more elegantly) and some sense of good structure would save a lot of work and be more elegant besides. Other little things might be useful.
It's really handy to write out your structure on paper before you start - are you familiar with finite state automata? - You want diagrams that look a bit like those. It helps. It also helps to know what you will get your net to do, and how you will teach it - again this comes down to structure, and courseware is really handy for providing examples. Here at UBC the course is at the 400 level, and I think the textbook is cited online. I will try and remember to attach a few links here when orkut is not being overrun by rabid monkeys.
However, I do agree that since I am human, the concept that I use for both consciousness and intelligence are that of a human. By consciousness I mean to be aware that I exist and also (this should not be part of the definition, but it helps) that I can think by myself.
The questions remain:
1) Is it possible to have human-like (or better than human) intelligence in other kind of hardware, a computer for example?
2) Is there a relationship, even some kind of cause and effect relationship between human-like intelligence and consciousness?
3) Very important but extremely hard do formalize. I do not mean simulation of "appearance" of intelligence in any way. I want true intelligence, whatever that means.
Friday, December 5, 2008
Hi people I've a question in my mind. It may be awkward...but still I want to discuss it. Normally what intelligence is defined is the way one act in certain situation but if we take a broader view then different things affect on ones intelligence that is natural ingenuity, creativity, knowledge and emotions or u can say emotional intelligence. If we take all these things together then we see that computer cannot act as human being. Computer is just mathematical methods which are discovered by human beings and if some genius discovers more powerful method then computer will fail. And it is true in many times. Then y we and other people compare human being and computer in this respect. It is obvious is not it CREATOR IS ALWAYS BEST THAN CREATION
As I mentioned computer cannot have the fundamental creativity. And creativity is defined as non algorithmic and discontinuous according to idealist theorists, example Abraham Maslow. There are two kinds of creative acts. The first, called situational creativity, is algorithmic and consists of inventive permutation and combinations of existing contexts to bring about something novel. This is the creativity that human most often pursue in their ego modality and computers can handle in principle. But there is also fundamental creativity which is a discovery of something new in a new context. This is the creativity that we ascribe to the great deeds of great people such as Einstein and Mozart, Tagore and Darwin. This kind of creativity involves the primary processes of the quantum self-experience. It is available to all of us but is relatively rare because it requires a quantum leap of discontinuity, transcendence, and freedom.
Like all Computer Science degree holders, I took an AI programming class. In it, we learned about various neural networks, designed to be similar to animal brains. All these were simulated in matrices for speed. I was toying with the idea of a neural net that simulated more features of the human brain. Things like cell death and new cell birth (part of forgetting and learning as resources are limited), overall system depression/happiness (for guiding learning in certain directions), dendrites of various lengths with stimulants to trigger growth (idea that neuron should connect to cells near them, forming chains and clusters of related concepts that then connect outwards). Anyway, as I played with all these, the matrix was getting more and more crowded. Besides, as a C/C++ programmer I find structs and objects more intuitive. I was wondering what people thought of the idea of simulating natural brains using a struct or object describing all these behaviors, occasionally spawning and killing off new ones, and having signals traverse the web of structs from the input points.
Thursday, December 4, 2008
With ever-increasing memory capacities and processing speeds and advances in distributed computation, and *assuming* that technological/scientific advances spur new understanding of the human brain in the next 50 years, I believe it will be possible within our lifetimes that we will have enough horsepower to emulate, to arbitrary precision as the years go on, a human brain. If this emulation is "good enough", then whether it will have a "self" is a purely philosophical issue. However, it seems certain that with enough understanding and horsepower, we can create intelligence.