Building Biological Neural Networks

Posted 17 May 2005 at 16:07 UTC by steve

Carbon nanotubes have been proposed as a replacement for silicon transistors but some researchers think a better long term replacement would be growing our own biological neural networks. The first step is to characterize the networks so we can design circuits and computers out of them. To that end, researchers Hugues Berry and Olivier Temam of INRIA have released a new paper titled, "haracterizing Self-Developing Biological Neural Networks: A First Step Towards their Application To Computing Systems" (PDF format). Among the potential problems of biological neurons: they're slow, many are faulty and they are hard to assemble. But they have one overriding advantage: we already know for certain that it's possible for neurons to self-assemble into fast, powerful, thinking machines.

Building Artificial Minds, posted 18 May 2005 at 05:14 UTC by AI4U » (Observer)

A Theory of Cognitivity explains how to transform neuronal networks into the artificial mind of the free on-line AI4U textbook of artificial intelligence.

practise what you preach, posted 18 May 2005 at 12:45 UTC by c6jones720 » (Master)

Mentifex I dont want to be rude, but why dont you practise what you preach and make something that actually works?

preaching the mindspel, posted 18 May 2005 at 13:32 UTC by AI4U » (Observer)

It is true that Mind.Forth has some problems in keeping its associations on track, but the situation is improving. In recent updates I have put a cap, a limit, on the activation-level of all the concepts engaged in Spreading- Activation to generate a thought. Now I need to fine-tune the AI Mind operation within the new limits. -Arthur (Mentifex)

Heretics & False gods, posted 18 May 2005 at 14:58 UTC by The Swirling Brain » (Master)

Is "haracterizing" the act of being a AI heretic character?

Those that espouse that they "already know for certain" how to do AI but can't do it, can't prove it, and have nothing to show for it, yet just babble on about it are in my opinion just nonsensical false AI gods. How can you claim to be an expert in something that you can't at all do? That's insanity and ignorance and meglomania! I suppose AI <u>is</u> the act of getting your head examined, but this is rediculous! It is heretical to abomnibal and perhaps even unethical to urge people to follow your plan toward AI that you yourself can't prove. Pity the blind fool that follows the blind fool when they both fall into the pit. Stupidity breads stupidity. These people should go live in Missouri (the "show-me" state) and learn what it means to show by example, not by unprovable theories.

unproven theories, posted 18 May 2005 at 15:51 UTC by steve » (Master)

On the other hand, without unproven theories, there'd be no AI news to report on! ;-)

But I get a little tired of the constant AI4U spamming too. If the robots.net community as a whole gets tired of it, they'll just remove their certifications of him (or remove their certs from those who certified him) and the problem will go away. No sign of that happening yet.

See more of the latest robot news!