A high performance floating-point special function unit using constrained piecewise quadratic approximation.
Davide De CaroNicola PetraAntonio G. M. StrolloPublished in: ISCAS (2008)
Keyphrases
- floating point
- taylor series expansion
- piecewise linear
- fixed point
- square root
- linear model
- instruction set
- sparse matrices
- floating point arithmetic
- graphics processing units
- convex functions
- smooth functions
- objective function
- fast fourier transform
- linear functions
- markov random field
- spline interpolation
- pairwise