Exploring the Precision Limitation for RRAM-Based Analog Approximate Computing.
IEEE Design & Test（2016）
摘要
RRAM-based approximate computing systems are significantly more energy efficient than many digital approximate computing systems, but their accuracy is less easily controlled and quantified. This paper analyzes the precision limitation for such systems and highlights the importance of RRAM device resolution in low-resistance states.
更多查看译文
关键词
Approximation methods,Approximate computing,Random access memory,Design automation,Signal resolution,Artificial neural networks,Performance evaluation
AI 理解论文
溯源树
样例
生成溯源树，研究论文发展脉络
Chat Paper
正在生成论文摘要