4 claims in “neuroscience, machine learning”
The backpropagation algorithm, while remarkably successful in machine learning, contradicts essential biological principles of brain function, making its exact implementation in neural tissue virtually impossible.
As neuroscience and artificial intelligence continue to inform each other, predictive coding stands as a compelling bridge between the remarkable learning capabilities of biological brains and the next generation of neural network architectures.
The brain is a gigantic ensemble of small models tailored to solve the ever-escalating number of tiny problems encountered; large brain functions can be decomposed into these specialized models, just as large neural networks contain extractable small models for specific tasks.
All neural models I've developed can operate autonomously and learn either unsupervised (learning from inputs alone) or supervised (learning with feedback about predictive success); perceptual and cognitive processes are designed to be general-purpose—they can respond adaptively to any input pattern in their domain and to changing environments.