Researchers from University of Notre Dame, Georgia Institute of Technology, and Villanova University published a technical paper titled “Probabilistic Memory for Trustworthy Edge Intelligence.”
Summary: The paper introduces p-MEM as “a unified memory primitive” that samples at “the native memory bandwidth.” It reports reductions in instruction count, sampling latency, and energy for Bayesian neural network workloads.
Find the technical paper here. July 2026.
Pei, Likai, Jiahao Zheng, Xueji Zhao, Emilie Ye, Jianbo Liu, Hanqing Tao, Ming-Yen Lee, Ruiyang Qin, Yiyu Shi, Shimeng Yu, X. Sharon Hu, and Ningyuan Cao. “Probabilistic Memory for Trustworthy Edge Intelligence.” arXiv, July 2026. Accepted for publication in the proceedings of the ACM/IEEE Design Automation Conference (DAC), 2026. https://doi.org/10.48550/arXiv.2607.02465.
Leave a Reply