The viability of analog-based accelerators for neuromorphic computing: a survey

M Musisi-Nkambwe, S Afshari, H Barnaby… - Neuromorphic …, 2021 - iopscience.iop.org
Focus in deep neural network hardware research for reducing latencies of memory fetches
has steered in the direction of analog-based artificial neural networks (ANN). The promise of …

A 40-nm MLC-RRAM compute-in-memory macro with sparsity control, on-chip write-verify, and temperature-independent ADC references

W Li, X Sun, S Huang, H Jiang… - IEEE Journal of Solid …, 2022 - ieeexplore.ieee.org
Resistive random access memory (RRAM)-based compute-in-memory (CIM) has shown
great potential for accelerating deep neural network (DNN) inference. However, device …

Nonvolatile Capacitive Crossbar Array for In‐Memory Computing

J Hur, YC Luo, A Lu, TH Wang, S Li… - Advanced Intelligent …, 2022 - Wiley Online Library
Conventional resistive crossbar array for in‐memory computing suffers from high static
current/power, serious IR drop, and sneak paths. In contrast, the “capacitive” crossbar array …

Nonvolatile capacitive synapse: device candidates for charge domain compute-in-memory

S Yu, YC Luo, TH Kim, O Phadke - IEEE Electron Devices …, 2023 - ieeexplore.ieee.org
Compute-in-memory (CIM) has emerged as a compelling approach to address the ever-
increasing demand for energy-efficient computing for edge artificial intelligence (AI) …

VSDCA: A voltage sensing differential column architecture based on 1T2R RRAM array for computing-in-memory accelerators

Z **g, B Yan, Y Yang, R Huang - IEEE Transactions on …, 2022 - ieeexplore.ieee.org
Non-volatile memory (NVM) such as RRAM and PCM has become the key component in
high energy efficiency computing-in-memory (CIM) architectures. However, the computing …

H3datten: Heterogeneous 3-d integrated hybrid analog and digital compute-in-memory accelerator for vision transformer self-attention

W Li, M Manley, J Read, A Kaul… - IEEE Transactions on …, 2023 - ieeexplore.ieee.org
After the success of the transformer networks on natural language processing (NLP), the
application of transformers to computer vision (CV) has followed suit to deliver …

Role of the electrolyte layer in CMOS-compatible and oxide-based vertical three-terminal ECRAM

G Han, J Seo, H Kim, D Lee - Journal of Materials Chemistry C, 2023 - pubs.rsc.org
Structured three-terminal electrochemical random access memory (3T-ECRAM) is
developed as a synaptic device at wafer scale using CMOS fabrication-compatible …

OCC: An automated end-to-end machine learning optimizing compiler for computing-in-memory

A Siemieniuk, L Chelini, AA Khan… - … on Computer-Aided …, 2021 - ieeexplore.ieee.org
Memristive devices promise an alternative approach toward non-Von Neumann
architectures, where specific computational tasks are performed within the memory devices …

Memory-immersed collaborative digitization for area-efficient compute-in-memory deep learning

S Nasrin, MB Hashem, N Darabi… - 2023 IEEE 5th …, 2023 - ieeexplore.ieee.org
This work discusses memory-immersed collaborative digitization among compute-in-
memory (CiM) arrays to minimize the area overheads of a conventional analog-to-digital …

ENNA: An efficient neural network accelerator design based on ADC-free compute-in-memory subarrays

H Jiang, S Huang, W Li, S Yu - IEEE Transactions on Circuits …, 2022 - ieeexplore.ieee.org
Compute-in-memory (CIM) is an attractive solution for machine learning hardware
acceleration since it merges computation directly into memory arrays, performing parallel …