Sparse Matrix to Matrix Multiplication: A Representation and Architecture for Acceleration | IEEE Conference Publication | IEEE Xplore