Thanks for presenting! Though I am not familiar with the field, the topic of studying the brain to alleviate "catastrophic forgetting" of ANN is interesting. In XdG, networks are split into several subnetworks a prior, I am wondering if layers/area of networks are identical or symmetric so that if multiples tasks of different types are studied/trained sequentially, the allocation of subnetworks to different tasks does not matter.
Thanks for presenting! Though I am not familiar with the field, the topic of studying the brain to alleviate "catastrophic forgetting" of ANN is interesting. In XdG, networks are split into several subnetworks a prior, I am wondering if layers/area of networks are identical or symmetric so that if multiples tasks of different types are studied/trained sequentially, the allocation of subnetworks to different tasks does not matter.