Research progress on memristor: From synapses to computing systems

X Yang, B Taylor, A Wu, Y Chen… - IEEE Transactions on …, 2022 - ieeexplore.ieee.org
As the limits of transistor technology are approached, feature size in integrated circuit
transistors has been reduced very near to the minimum physically-realizable channel length …

[HTML][HTML] Toolflow for the algorithm-hardware co-design of memristive ANN accelerators

M Wabnitz, T Gemmeke - Memories-Materials, Devices, Circuits and …, 2023 - Elsevier
The capabilities of artificial neural networks are rapidly evolving, so are the expectations for
them to solve ever more challenging tasks in numerous everyday situations. Larger, more …

Architecting decentralization and customizability in dnn accelerators for hardware defect adaptation

E Ozen, A Orailoglu - … on Computer-Aided Design of Integrated …, 2022 - ieeexplore.ieee.org
The efficiency of machine intelligence techniques has improved noticeably in the embedded
application domains thanks to the dedicated hardware accelerators for deep neural …

TEFLON: Thermally Efficient Dataflow-Aware 3D NoC for Accelerating CNN Inferencing on Manycore PIM Architectures

G Narang, C Ogbogu, JR Doppa… - ACM Transactions on …, 2024 - dl.acm.org
Resistive random-access memory (ReRAM)-based processing-in-memory (PIM)
architectures are used extensively to accelerate inferencing/training with convolutional …

Hybrid RRAM/SRAM in-memory computing for robust DNN acceleration

G Krishnan, Z Wang, I Yeo, L Yang… - … on Computer-Aided …, 2022 - ieeexplore.ieee.org
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 …

Approximate computing and the efficient machine learning expedition

J Henkel, H Li, A Raghunathan, MB Tahoori… - Proceedings of the 41st …, 2022 - dl.acm.org
Approximate computing (AxC) has been long accepted as a design alternative for efficient
system implementation at the cost of relaxed accuracy requirements. Despite the AxC …

Essence: Exploiting structured stochastic gradient pruning for endurance-aware reram-based in-memory training systems

X Yang, H Yang, JR Doppa, PP Pande… - … on Computer-Aided …, 2022 - ieeexplore.ieee.org
Processing-in-memory (PIM) enables energy-efficient deployment of convolutional neural
networks (CNNs) from edge to cloud. Resistive random-access memory (ReRAM) is one of …

Harmonica: Hybrid Accelerator to Overcome Imperfections of Mixed-signal DNN Accelerators

P Behnam, U Kamal, A Shafiee… - 2024 IEEE …, 2024 - ieeexplore.ieee.org
In recent years, PIM-based mixed-signal accelerators have been proposed as energy-and
area-efficient solutions with ultra-high throughput to accelerate DNN computations …

Neuro-Symbolic Computing: Advancements and Challenges in Hardware-Software Co-Design

X Yang, Z Wang, XS Hu, CH Kim, S Yu… - … on Circuits and …, 2023 - ieeexplore.ieee.org
The rapid progress of artificial intelligence (AI) has led to the emergence of a highly
promising field known as neuro-symbolic (NeSy) computing. This approach combines the …

TransCODE: Co-design of transformers and accelerators for efficient training and inference

S Tuli, NK Jha - IEEE Transactions on Computer-Aided Design …, 2023 - ieeexplore.ieee.org
Automated co-design of machine learning models and evaluation hardware is critical for
efficiently deploying such models at scale. Despite the state-of-the-art performance of …