作者: Jie Han , Michael Orshansky
关键词:
摘要: Approximate computing has recently emerged as a promising approach to energy-efficient design of digital systems. relies on the ability many systems and applications tolerate some loss quality or optimality in computed result. By relaxing need for fully precise completely deterministic operations, approximate techniques allow substantially improved energy efficiency. This paper reviews recent progress area, including arithmetic blocks, pertinent error measures, algorithm-level computing.