Fernando Aguirre; Abu Sebastian; Manuel Le Gallo; Wenhao Song; Tong Wang; J. Joshua Yang; Wei Lü; Meng‐Fan Chang; Daniele Ielmini; Yuchao Yang; Adnan Mehonić; Anthony J. Kenyon; Marco A. Villena; J.B. Roldán; Yuting Wu; Hung-Hsi Hsu; Nagarajan Raghavan; J. Suñé; E. Miranda; Ahmed M. Eltawil; Gianluca Setti; Kamilya Smagulova; K. Saláma; Olga Krestinskaya; Xiaobing Yan; Kah‐Wee Ang; Samarth Jain; Sifan Li; Osamah Alharbi; Sebastián Pazos; Mario Lanza · 2024 · Nature Communications
Paper
Artificial Intelligence (AI) is currently experiencing a bloom driven by deep learning (DL) techniques, which rely on networks of connected simple computing units operating in parallel. The low communication bandwidth between memory and processing units in conventional von Neumann machines does not support the requirements of emerging applications that rely extensively on large sets of data. More recent computing paradigms, such as high parallelization and near-memory computing, help alleviate the data communication bottleneck to some extent, but paradigm- shifting concepts are required. Memristors, a novel beyond-complementary metal-oxide-semiconductor (CMOS) technology, are a promising choice for memory devices due to their unique intrinsic device-level properties, enabling both storing and computing with a small, massively-parallel footprint at low power. Theoretically, this directly translates to a major boost in energy efficiency and computational throughput, but various practical challenges remain. In this work we review the latest efforts for achieving hardware-based memristive artificial neural networks (ANNs), describing with detail the working principia of each block and the different design alternatives with their own advantages and disadvantages, as well as the tools required for accurate estimation of performance metrics. Ultimately, we aim to provide a comprehensive protocol of the materials and methods involved in memristive neural networks to those aiming to start working in this field and the experts looking for a holistic approach.
Analysis
This paper reviews the latest efforts in hardware-based memristive artificial neural networks (ANNs), detailing their principles, design alternatives, and performance estimation tools to guide researchers in the field.
Discovery
Alex James
Emmanuel Joseph Shaji; Zhiyuan Li; Srikanth Doddapaneni; Bhavya Rakheja; Vikrant Chaudhary; Avantika Suthar; Lingyun Zhu; Jingxin Ma; Hongbin Zhang; Monojit Bag; Gerardo Hernandez-Sosa; Ramesh Kumar
Falihah Balqis; Jin Pyo Lee; Zhenxiang Xing; Hui Wang; Tupei Chen; Rong Ji; Pooi See Lee
Hanna Hamrell; Thomas Sjögren; Hannes Ovrén
J. Rajvel
Minsu Ko; Yongjin Byun; Sungjun Kim
Source record