DISTRIBUTED CELLULAR COMPUTING SYSTEM AND METHOD FOR NEURAL-BASED SELF-ORGANIZING MAPS

摘要

The present invention provides a neuromorphic computing system configured to be trained using unsupervised learning through distributed computing means. The neuromorphic computing system comprises an artificial neural network implemented as a grid of locally connected cells wherein each cell comprises hardware components for neural computing and storage, and is connected to its direct closest neighbors. The neuromorphic computing system comprises a clock system providing periodic active clock edges allowing in each cell to simultaneously and synchronously compute the neuron's Euclidean distance to the input, then compute the Best Matching Unit and the Manhattan distance to it in multiple clock cycles based on a time to Manhattan distance transformation, and finally update the neuron's weights. Advantageously, the iterative method of the present invention brings a formalized, validated, generic and hardware-efficient solution to the scalability problem of centralized and fully- connected distributed SOMs implementations. The system operates with the same clock frequency regardless of the number of neurons, such that the input rate evolves in square root complexity with respect to the number of neurons in the grid.