[图书][B] Efficient processing of deep neural networks
This book provides a structured treatment of the key principles and techniques for enabling
efficient processing of deep neural networks (DNNs). DNNs are currently widely used for …
efficient processing of deep neural networks (DNNs). DNNs are currently widely used for …
Mixed-signal computing for deep neural network inference
B Murmann - IEEE Transactions on Very Large Scale …, 2020 - ieeexplore.ieee.org
Modern deep neural networks (DNNs) require billions of multiply-accumulate operations per
inference. Given that these computations demand relatively low precision, it is feasible to …
inference. Given that these computations demand relatively low precision, it is feasible to …
Machine learning at the edge
M Verhelst, B Murmann - NANO-CHIPS 2030: On-Chip AI for an Efficient …, 2020 - Springer
Abstract Machine learning algorithms have been successfully deployed in cloud-centric
applications and on computationally powerful digital platforms such as high-end FPGAs and …
applications and on computationally powerful digital platforms such as high-end FPGAs and …
A 4-bit mixed-signal MAC array with swing enhancement and local kernel memory
Modern deep neural networks require energy-and area-efficient multi-bit multiply-
accumulate (MAC) functions. In-memory computing (IMC) with analog accumulation has …
accumulate (MAC) functions. In-memory computing (IMC) with analog accumulation has …
A Multi-Bit Near-RRAM based Computing Macro with Highly Computing Parallelism for CNN Application
KC Lin, H Zuo, HY Wang, YP Huang… - … , Automation & Test …, 2024 - ieeexplore.ieee.org
Resistive random-access memory (RRAM) based compute-in-memory (CIM) is an emerging
approach to address the demand for practical implementation of artificial intelligence (AI) on …
approach to address the demand for practical implementation of artificial intelligence (AI) on …
Read only memory architecture for analog matrix operations
E Papageorgiou, K Wojciechowski… - US Patent …, 2022 - Google Patents
A read-only memory (ROM) computing unit utilized in matrix operations of a neural network
comprising a unit element including one or more connections, wherein a weight associated …
comprising a unit element including one or more connections, wherein a weight associated …
Exploration of security threats in In-Memory Computing Paradigms
P Inglese - 2023 - theses.hal.science
Computation in Memory (CIM) is a groundbreaking concept that involves performing
computations directly within the memory itself, eliminating the need to transfer data back and …
computations directly within the memory itself, eliminating the need to transfer data back and …
Digital to Pulse Converter for Analog in Memory Compute Applications
SB Naik, A Iqbal - 2023 IEEE Asia Pacific Conference on …, 2023 - ieeexplore.ieee.org
This work presents a novel approach in Digital to Pulse Converter (DPC) design, which is
used as an alternative to Digital to Analog Converter (DAC) for Analog in Memory Compute …
used as an alternative to Digital to Analog Converter (DAC) for Analog in Memory Compute …
Mixed-Signal Compute and Memory Fabrics for Deep Neural Networks
B Murmann - Analog Circuits for Machine Learning, Current/Voltage …, 2022 - Springer
Deep neural networks have emerged as important new drivers for custom VLSI computing
hardware in resource-constrained environments. In this context, this chapter reviews …
hardware in resource-constrained environments. In this context, this chapter reviews …
[图书][B] RRAM Compute-in-Memory Hardware for Efficient, Versatile, and Accurate ai Inference
W Wan - 2022 - search.proquest.com
Performing ever-demanding artificial intelligence (AI) tasks directly on resource-constrained
edge devices calls for unprecedented energy-efficiency of edge AI hardware. AI hardware …
edge devices calls for unprecedented energy-efficiency of edge AI hardware. AI hardware …