CHIMERA: A 0.92-TOPS, 2.2-TOPS/W Edge AI Accelerator With 2-MByte On-Chip Foundry Resistive RAM for Efficient Training and Inference
Author:
Affiliation:
1. Department of Electrical Engineering, Stanford University, Stanford, CA, USA
2. Taiwan Semiconductor Manufacturing Company (TSMC), Hsinchu, Taiwan
3. Reality Labs, Meta Platforms, Inc, Sunnyvale, CA, USA
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Subject
Electrical and Electronic Engineering
Link
https://ieeexplore.ieee.org/ielam/4/9741814/9690772-aam.pdf
Reference67 articles.
1. A high-performance low-power highly manufacturable embedded DRAM technology using backend hi-$k$ MIM capacitor at 40 nm node and beyond;chen;Proc Int Symp VLSI Technol Syst Appl,2011
2. A 40 nm split gate embedded flash macro with flexible 2-in-1 architecture, code memory with 140 MHz read speed and data memory with 1M cycles endurance;yu;Proc Symp VLSI Circuits,2017
3. 7.4 GANPU: A 135TFLOPS/W Multi-DNN Training Processor for GANs with Speculative Dual-Sparsity Exploitation
4. Illusion of large on-chip memory by networked computing chips for neural network inference
5. CHIMERA: A 0.92 TOPS, 2.2 TOPS/W Edge AI Accelerator with 2 MByte On-Chip Foundry Resistive RAM for Efficient Training and Inference
Cited by 14 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Write–Verify-Free MLC RRAM Using Nonbinary Encoding for AI Weight Storage at the Edge;IEEE Transactions on Very Large Scale Integration (VLSI) Systems;2024-02
2. Design principles for lifelong learning AI accelerators;Nature Electronics;2023-11-16
3. A Low Power AI Hardware Accelerator for Microwave-Based Ice Detection;2023 IEEE SENSORS;2023-10-29
4. PBA: Percentile-Based Level Allocation for Multiple-Bits-Per-Cell RRAM;2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD);2023-10-28
5. Stepped Identical Reset-Pulse and Lookup Table Programming of Multilevel Conductance in HfO x /AlO y Superlattice-Like Memristors;IEEE Transactions on Electron Devices;2023-09
1.学者识别学者识别
2.学术分析学术分析
3.人才评估人才评估
"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370
www.globalauthorid.com
TOP
Copyright © 2019-2024 北京同舟云网络信息技术有限公司 京公网安备11010802033243号 京ICP备18003416号-3