statementSaid on 2026-10-31
Kempa presents an opposite approach called NEF (Neural Engineering Framework): if we already know the mathematical formula, the network does not need to learn it from scratch; instead a simple equation is solved once to find the exact weights, compiling formulas directly into the network with zero training.
A statement by author, said on 2026-10-31. Statements are as reported by the cited publishers — not verified by DecisionGateway.
קמפה מציג גישה הפוכה לגמרי שנקראת NEF (שזה Neural Engineering Framework): אם אנחנו כבר יודעים את הנוסחה המתמטית, למה שהרשת תלמד אותה מאפס?
Discussion
Related claims
In a deep and very enjoyable review, Volker Kempa presents the alternative gaining engineering momentum: Neuromorphic AI based on Spiking Neural Networks (SNNs).author · statementIn a regular network, if the model errs, the backprop algorithm stops everything, goes back from end to start and updates all the weights; in the brain there is no such central manager, and instead each synapse learns on its own from its local environment using STDP.author · statementEach implanted layer is multiplied by a learned gate initialized to zero; at the moment of implantation the network does not change at all, and the gradient descent decides whether to open the gate.post author · statementThe researchers show that training leads the model to learn an efficient solution path, even though it receives no extra reward for choosing that path.החוקרים · investigationAccording to the assessment of the apparatus leadership, using neural networks will cost the state less than hiring additional employees to perform a comparable volume of work.оценке руководства аппарата · predictionAccording to the assessment of the apparatus leadership, using neural networks will cost the state less than hiring additional employees to perform a comparable volume of work.Руководство аппарата · prediction