3 ms·
"Were guessing plausible ways the brain might work in ways similar to the machine learning world, but have no data to back it up"
by londons_explore 7y ago
"Were guessing plausible ways the brain might work in ways similar to the machine learning world, but have no data to back it up"
- GChevalier 7y ago> "The relationship of spike-time-dependent plasticity to other models requires further clarifying work" - I got something for that, in fact I think I discovered that backpropagation engenders STDP: https://github.com/guillaume-chevalier/Spiking-Neural-Network-SNN-with-PyTorch-where-Backpropagation-engenders-STDP https://github.com/guillaume-chevalier/Spiking-Neural-Networ...
- ahartmetz 7y agoBackpropagation feels like a pretty gross hack to me anyway. It is a powerful component that still needs a large amount of additional hand-engineering to do more than recognize shapes it has seen before. And it's quite mathematical, which seems kind of difficult to implement in a brain. Edit: The article and code try to show how backpropagation might work given neuron behavior. I'll leave discussing the quality of the model to biologists.
- marmaduke 7y ago> And it's quite mathematical, which seems kind of difficult to implement in a brain. Another commenter pointed out that you don't need perfect backprop, some random backprijections are sufficient.