4 ms·
Unfortunately, we do not know how neurons learn, compute and represent information in biological neural networks. Though I feel that there would be great potent
by IdiocyInAction 6y ago
Unfortunately, we do not know how neurons learn, compute and represent information in biological neural networks. Though I feel that there would be great potential in figuring this out - the energy efficiency and capabilities of the brain are truly amazing.
- Balgair 6y agoHuh? >Learn Depends on the neuron, but things like NMDA and AMPA receptors are very well studied at this point and are the primary portion of synaptic plasticity, again, depending on the neuron: https://en.wikipedia.org/wiki/AMPA_receptor https://en.wikipedia.org/wiki/AMPA_receptor >Compute Again, depends on the neuron, but dendritic integration is a classic example: https://www.sciencedirect.com/topics/neuroscience/dendritic-integration https://www.sciencedirect.com/topics/neuroscience/dendritic-... > Represent information The do so in the frequency space. But again, it's highly dependent on the system of neurons : https://en.wikipedia.org/wiki/Neural_coding https://en.wikipedia.org/wiki/Neural_coding That said, holy cow yes. The brain is really something when it comes to power usage.
- IdiocyInAction 6y agoNeural coding is not a solved problem. Rate codes are not sufficient to explain all experimental data. Neither is learning; we know some things, sure, we have models for long- and short-term plasticity, but we have no idea how learning actually works. With computation, there are many hypothesis and dendrites do indeed do more than you would think, but again, these are just small details.