Issue No. 03 - May-June (2013 vol. 33)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MM.2013.28
Hadi Esmaeilzadeh , University of Washington
Adrian Sampson , University of Washington
Luis Ceze , University of Washington
Doug Burger , Microsoft Research
This work proposes an approximate algorithmic transformation and a new class of accelerators, called neural processing units (NPUs). NPUs leverage the approximate algorithmic transformation that converts regions of code from a Von Neumann model to a neural model. NPUs achieve an average 2.3× speedup and 3.0× energy savings for general-purpose approximate programs. This new class of accelerators shows that significant performance and efficiency gains are possible when the abstraction of full accuracy is relaxed in general-purpose computing.
Computer architecture, Neural networks, Approximation algorithms, Algorithm design and analysis, Accelerators, NPUs, approximate computing, accelerators, neural networks, Parrot algorithmic transformation, neural processing units
D. Burger, L. Ceze, H. Esmaeilzadeh and A. Sampson, "Neural Acceleration for General-Purpose Approximate Programs," in IEEE Micro, vol. 33, no. , pp. 16-27, 2013.