Efficient Fpga Implementation of a Generic Function Approximator and Its Application to Neural Net Computation

doi 10.1109/mwscas.2003.1562418