Hybrid RRAM/SRAM in-memory computing for robust DNN acceleration
RRAM-based in-memory computing (IMC) effectively accelerates deep neural networks
(DNNs) and other machine learning algorithms. On the other hand, in the presence of RRAM …
(DNNs) and other machine learning algorithms. On the other hand, in the presence of RRAM …
Stochastic computing in beyond von-neumann era: Processing bit-streams in memristive memory
Stochastic Computing (SC) is an alternative computing paradigm that promises high
robustness to noise and outstanding area-and power-efficiency compared to traditional …
robustness to noise and outstanding area-and power-efficiency compared to traditional …
In-memory computing circuit implementation of complex-valued hopfield neural network for efficient portrait restoration
Complex-valued neural networks have better optimization capabilities, stronger robustness,
and richer characterization capabilities compared with real-valued neural networks, which …
and richer characterization capabilities compared with real-valued neural networks, which …
BC-MVLiM: A binary-compatible multi-valued logic-in-memory based on memristive crossbars
Y Sun, Z Li, W Liu, W He, Q Wang… - IEEE Transactions on …, 2023 - ieeexplore.ieee.org
Logic-in-memory with memristive crossbars is an attractive approach for realizing beyond
von Neumann architectures. Multi-valued logic (MVL) containing more than two logic levels …
von Neumann architectures. Multi-valued logic (MVL) containing more than two logic levels …
Training-free stuck-at fault mitigation for ReRAM-based deep learning accelerators
C Quan, ME Fouda, S Lee, G Jung… - … on Computer-Aided …, 2022 - ieeexplore.ieee.org
Although Resistive RAMs can support highly efficient matrix–vector multiplication, which is
very useful for machine learning and other applications, the nonideal behavior of hardware …
very useful for machine learning and other applications, the nonideal behavior of hardware …
A quantized convolutional neural network implemented with memristor for image denoising and recognition
The interference of noise will cause the degradation of image quality, which can have a
negative impact on the subsequent image processing and visual effect. Although the …
negative impact on the subsequent image processing and visual effect. Although the …
An Empirical Fault Vulnerability Exploration of ReRAM-Based Process-in-Memory CNN Accelerators
A Dorostkar, H Farbeh… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Resistive random-access memory (ReRAM)-based processing-in-memory (PIM) accelerator
is a promising platform for processing massively memory intensive matrix-vector …
is a promising platform for processing massively memory intensive matrix-vector …
RePAST: A ReRAM-based PIM Accelerator for Second-order Training of DNN
The second-order training methods can converge much faster than first-order optimizers in
DNN training. This is because the second-order training utilizes the inversion of the second …
DNN training. This is because the second-order training utilizes the inversion of the second …
Hill climbing for efficient spiking neural network acceleration on neuromorphic chips
Spiking neural networks (SNNs) emulate biological neurons transmitting information through
discrete spikes or pulses of activity. SNNs found extensive application in neuromorphic …
discrete spikes or pulses of activity. SNNs found extensive application in neuromorphic …
A Hardware Friendly Variation-Tolerant Framework for RRAM-Based Neuromorphic Computing
FY Gu, CH Yang, C Lin, DW Chang… - … on Circuits and …, 2024 - ieeexplore.ieee.org
Emerging resistive random access memory (RRAM) attracts considerable interest in
computing-in-memory by its high efficiency in multiply-accumulate operation, which is the …
computing-in-memory by its high efficiency in multiply-accumulate operation, which is the …