4.6 Article

STT-BSNN: An In-Memory Deep Binary Spiking Neural Network Based on STT-MRAM

相关参考文献

注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article Engineering, Electrical & Electronic

Fully On-Chip MAC at 14 nm Enabled by Accurate Row-Wise Programming of PCM-Based Weights and Parallel Vector-Transport in Duration-Format

P. Narayanan et al.

Summary: This article introduces a 14nm test chip for Analog AI inference, utilizing phase change memory devices to achieve high-efficiency MAC operations and DNN weight storage. With a closed-loop tuning algorithm, accurate weight programming and transfer are achieved, leading to near-software-equivalent accuracy on two different DNNs.

IEEE TRANSACTIONS ON ELECTRON DEVICES (2021)

Article Neurosciences

Neural Coding in Spiking Neural Networks: A Comparative Study for Robust Neuromorphic Systems

Wenzhe Guo et al.

Summary: This study compared four neural coding schemes and found that TTFS coding achieved the highest computational performance with low latency and fewer SOPs. Phase coding was most resilient to input noise, while burst coding showed the highest network compression efficacy and robustness to hardware non-idealities.

FRONTIERS IN NEUROSCIENCE (2021)

Article Computer Science, Artificial Intelligence

Deep Spiking Neural Networks With Binary Weights for Object Recognition

Yixuan Wang et al.

Summary: The study proposes an efficient weights-threshold balance conversion method to obtain SNNs with binary weights, resulting in a significant reduction in memory storage. Experimental results show that the binary SNN not only requires much less memory resources compared to its high-precision counterpart, but also achieves high-recognition accuracy comparable to other state-of-the-art SNNs.

IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS (2021)

Proceedings Paper Engineering, Electrical & Electronic

STT-MRAM Architecture with Parallel Accumulator for In-Memory Binary Neural Networks

Thi-Nhan Pham et al.

Summary: This paper presents a row-wise XNOR accumulator architecture for STT-MRAM arrays, which enables parallel and efficient multiply-and-accumulate operations, and is suitable for in-memory computing and binary neural network applications. The architecture, which does not require ADC, can achieve high classification accuracy for XNOR vector sizes of 128 bits.

2021 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS) (2021)

Proceedings Paper Engineering, Electrical & Electronic

VSA: Reconfigurable Vectorwise Spiking Neural Network Accelerator

Hong-Han Lien et al.

Summary: This work presents a binary weight spiking model with small time steps and low hardware cost for low-power design on edge devices. Additionally, a vectorwise hardware accelerator is proposed to reduce memory bandwidth and achieve competitive accuracy on the MNIST and CIFAR-10 datasets.

2021 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS) (2021)

Article Computer Science, Artificial Intelligence

A Probabilistic Synapse With Strained MTJs for Spiking Neural Networks

Samuel N. Pagliarini et al.

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2020)

Article Computer Science, Hardware & Architecture

Broad-Purpose In-Memory Computing for Signal Monitoring and Machine Learning Workloads

Saurabh Jain et al.

IEEE SOLID-STATE CIRCUITS LETTERS (2020)

Article Computer Science, Information Systems

Stochastic Computing for Hardware Implementation of Binarized Neural Networks

Tifenn Hirtzlin et al.

IEEE ACCESS (2019)

Article Engineering, Electrical & Electronic

An efficient ReRAM-based inference accelerator for convolutional neural networks via activation reuse

Yan Chen et al.

IEICE ELECTRONICS EXPRESS (2019)

Article Engineering, Electrical & Electronic

Eyeriss: An Energy-Efficient Reconfigurable Accelerator for Deep Convolutional Neural Networks

Yu-Hsin Chen et al.

IEEE JOURNAL OF SOLID-STATE CIRCUITS (2017)

Article Engineering, Electrical & Electronic

An Efficient Heterogeneous Memristive XNOR for In-Memory Computing

Muath Abu Lebdeh et al.

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS (2017)

Article Engineering, Electrical & Electronic

Proposal for a Leaky-Integrate-Fire Spiking Neuron Based on Magnetoelectric Switching of Ferromagnets

Akhilesh Jaiswal et al.

IEEE TRANSACTIONS ON ELECTRON DEVICES (2017)

Proceedings Paper Engineering, Electrical & Electronic

DNPU: An 8.1TOPS/W Reconfigurable CNN-RNN Processor for General-Purpose Deep Neural Networks

Dongjoo Shin et al.

2017 IEEE INTERNATIONAL SOLID-STATE CIRCUITS CONFERENCE (ISSCC) (2017)

Article Engineering, Electrical & Electronic

Novel Boosted-Voltage Sensing Scheme for Variation-Resilient STT-MRAM Read

Quang Kien Trinh et al.

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS (2016)

Article Engineering, Electrical & Electronic

Probabilistic Deep Spiking Neural Systems Enabled by Magnetic Tunnel Junction

Abhronil Sengupta et al.

IEEE TRANSACTIONS ON ELECTRON DEVICES (2016)

Article Engineering, Electrical & Electronic

Voltage Scaled STT-MRAMs Towards Minimum-Energy Write Access

Quang Kien Trinh et al.

IEEE JOURNAL ON EMERGING AND SELECTED TOPICS IN CIRCUITS AND SYSTEMS (2016)

Article Computer Science, Theory & Methods

A Survey Of Architectural Approaches for Managing Embedded DRAM and Non-Volatile On-Chip Caches

Sparsh Mittal et al.

IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS (2015)

Article Engineering, Electrical & Electronic

A 10-bit Charge-Redistribution ADC Consuming 1.9 μW at 1 MS/s

Michiel van Elzakker et al.

IEEE JOURNAL OF SOLID-STATE CIRCUITS (2010)

Article Engineering, Electrical & Electronic

Advances and Future Prospects of Spin-Transfer Torque Random Access Memory

E. Chen et al.

IEEE TRANSACTIONS ON MAGNETICS (2010)