Straying from the current trends in deep learning, there is an, arguably, interesting idea of neuronal ensembles possibly providing an alternative to the current "layered feature detectors" framework for neural network construction by being considered a basic computational unit instead of one feature detecting neuron. This idea certainly has at least some presence in neuroscience circles, but I found it hard to find studies, attempting to obtain a working computational ensemble based model, which would try to solve any of the existing computer vision/NLP tasks or anything of the sort. This may be just due to me looking in the wrong places, but in any case, I would appreciate any references to papers, exploring building neural network architectures with neuronal ensemblies involvement.

Just to be clear, I would be interested in any papers on computational modelling of ensemblies even if they are not trying to solve any particular ML task, but it would be better, if the topic of the research is more closely aligned with computer science instead of neurobiology even if CS connection is of a more exotic kind; for example, paper trying to see, whether you can store different concepts and their relations in the ensemble based network is more desirable than paper, trying to accurately model individual neuron and synaptic plasticity dynamics and see, that ensembles emerge, if you scale the system. But again, I would be glad to get references to research in both of these example topics and many more.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.