Geek Answers Handbook

Human brain emulation?

Is there any open source project that is trying to implement and emulate the human brain and feelings in computer software?

I believe in the future technology of artificial intelligence, and I always wanted to contribute to this technology and learn about it.

Thanks.

+7
artificial-intelligence open-source
source share
9 answers

We do not know enough about how the brain works to try to do what you say fundamentally. (That is, something like "fading wildly"). So this is not a software issue - if we had any idea of ​​what to write, perhaps it will be, but now we are not doing it.

However, you may be interested in the Blue Brain Project for a more biological approach or in any of several machine learning projects like the DARPA Autonomous Vehicle Grand Challenge . A less useful but more conversational approach can be found in ALICE , but I would not recommend anything useful.

+6
source share

Using the brain for over 50 years, this is the last thing I would choose for AI modeling. Brains are known to be unreliable and arbitrary, and have hidden biases that may require a reduction in years.

+7
source share

Jeff Hawkins, author of On Intelligence, owns Numenta. He has a theory about how brains work, and Numenta has a software product that models it. I downloaded and played with it a while ago, it seems to be very good at image recognition. I'm not quite sure what licensing is, I think it's free for academic purposes. In addition, the website is currently down.

http://en.wikipedia.org/wiki/Numenta

http://www.numenta.com/

+7
source share

Most of the AI ​​lectures I took at school were professors who have been chasing the dream of “strong AI” for many years, and finally realized that if they could barely understand how the human brain and the functions of the mind (and theories behind these functions sometimes changing almost daily), how can they ever hope to imitate this artificially? Most of them came to terms with AI in niches where this problem is more clearly defined: first-time walkthrough, application of SAT-solution, image processing, chess awards, conversation, etc ... but they gave on the true, general purpose of the “thinking machine” "

My advice would be to study a specific problem: you are interested (for example, finding paths, SAT-solver applications, such as diagnostic systems, etc.), and see what AI approaches were taken to solve them, Perhaps the problem that you interests, does not have much in terms of solutions for AI. In this case, you can start work on a new one!;)

... But you probably have to narrow it down to a certain class of problems if you do not want to be overwhelmed - at least at first.

+2
source share

The field you are looking for is Machine Learning . In particular, evolutionary algorithms such as genetic algorithms or genetic programming. One of the algorithms that I know, specifically designed to simulate the human brain, is the Hierarchical Temporary Memory, which I read about here . But this is a very complex problem, and we were still FLYING from in any way to comprehend the human brain.

+1
source share

There are algorithms that model the human brain. They are called artificial neural networks (ANNs). They mainly simulate synapses and try to simulate the way our synapses can receive signals, and if the combined signal input is strong enough, run your own signals along the dendrites to other synapses.

The fact is that creating ANN as a method of trying to imitate the real thing is very similar to using nuclear weapons to simulate the sun; Of course, this will give you valuable data, but from the point of view of his ability to approximate what he models, he falls WAYYY short.

I am not 100% positive on a relative scale here, but to give a decent idea, consider the following (it will definitely be canceled by several orders of magnitude ... but it is close enough to get an idea of ​​why ANN do not rule the world for us):

If you took every computer on the planet and used every available resource to create the largest ANNs they could, and then connected all these different ANNs to each other (thus creating an even larger ANN), you MAY start to approach the amount of compounds present in the human brain.

+1
source share

You can see Cyc :

Cyc is an artificial intelligence project that is trying to collect a comprehensive ontology and knowledge, the foundation of everyday common sense knowledge in order to provide an AI application for performing human-like reasoning. The project was started in 1984 by Douglas Lenat at MCC and is developed by Cycorp. Parts of the project are released as OpenCyc, which provides an API, an RDF endpoint, and an open source data dump data license.

Not exactly the brain, but an important component of intelligent intelligence.

0
source share

There is a field of computer science known as Organic Computing http://en.wikipedia.org/wiki/Organic_computing Some of the goals of these efforts are as follows.

  • self-organization
  • self-configuration (automatic configuration)
  • self-optimization (automatic optimization)
  • self healing
  • self-defense (automated security computer)
  • self explanations
  • context awareness
0
source share

The nearest possible thing that I know will be the structure of the watchmaker . Although not related to the human brain, it seems to be striving for an AI type structure.

http://watchmaker.uncommons.org/

The Watchmaker Framework is an extensible, high-performance, object-oriented framework for implementing platform-independent evolutionary / genetic algorithms in Java

0
source share

More articles:


All Articles