In biologic systems coupling ionic and electronic signals is necessary for learning. Nowadays we try to pursue artificial intelligence employing only electronic signals, I wonder if this is possible or if we may somehow incorporate ionic and electronic signals in the system, for instance with organic electrochemical transistors.
Jonathan,
why do you want to implement ionic signal processing in artificial systems? What should be the benefit of this attempt?
I don't know, it might not be necessary. In the human brain they do are necessary.
The theory of neural networks was set forth by Warren McCulloch and Walter Pitts in the 1940s.
"Nature of the impulse. The impulse is electrical in nature. Ions contained within and around the neural membrane include sodium, potassium, and calcium, all of which are positively charged, and chloride, which is negatively charged. Before the impulse begins, the neuron is said to be in resting state or polarized. All the positive ions are outside the neural membrane, and all the negative ones are on the inside. To transmit the impulse, the ions cross over to reverse this pattern, resulting in a wave of depolarization. Once the impulse has passed, the neuron is completely depolarized, and it must be restored to a resting state to send a new impulse..."
Some neural systems studied in artificial intelligence try to mimic this process. My question is that... is all this really necessary? Maybe we can just learn with 100% pure electronic signals, what do you think? On the other hand, computers and human brain work fundamentally different so far, "intelligent supercomputers" simply compute thousands of operations and try to come with a best solution after a mathematical model, while we humans we create models from our stimulus with nature and other humans and we are more intuitive. Intuition usually is associated with emotions. Human brain is amazing, it weights about 1.3 kg, and consumes about 20 Watts... I would like to learn more about it. Thanks for your comments. I appreciate it.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Artificial Intelligence