A General-Purpose Method for Faithfully Rounded Floating-Point Function Approximation in FPGAs | IEEE Conference Publication | IEEE Xplore