Hardware enhancements to radial basis function with restricted coulomb energy learning and/or k-nearest neighbor based neural network classifiers
Assigneein2H2
InventorsBruce K. McCormick, BILL H. NAGEL, CHRIS J. MCCORMICK, MICK FANDRICH
Life of this patent
Estimating the term…
Abstract
This disclosure describes embodiments for a hardware based neural network integrated circuit classifier incorporating natively implemented Radial Basis functions, Restricted Coulomb Energy function, and/or kNN to make it more practical for handling a broader group of parallel algorithms
Claims25
A nonlinear neuron classifier comprising a neuron array that processes an input vector of variable component length with variable component depths to be input into the classifier for comparison against vectors already stored or learned in the classifier, a system bus interface for detecting and capturing the input vector, and logic that is connected to individual neurons in parallel that broadcasts the input vector concurrently to vectors of similar length and depth stored in the classifier neurons, for calculation.
Dependencies are inferred from the claim text.