SPX: Collaborative Research: Ula! - An Integrated Deep Neural Network (DNN) Acceleration Framework with Enhanced Unsupervised Learning Capability
SPX: Collaborative Research: Ula! - An Integrated Deep Neural Network (DNN) Acceleration Framework with Enhanced Unsupervised Learning Capability
批准号:
1725447
负责人:
Yuan Xie
金额:
$28.0万
依托单位国家:
美国
项目类别:
Standard Grant
财政年份:
2017
资助国家:
美国
项目状态:
已结题
起止时间:
2017-09-01 至 2021-08-31
中文摘要
点击翻译按钮获取中文摘要
英文摘要
In light of very recent revolutions of unsupervised learning algorithms (e.g., generative adversarial networks and dual-learning) and the emergence of their applications, three PIs/co-PI from Duke and UCSB form a team to design Ula! - an integrated DNN acceleration framework with enhanced unsupervised learning capability. The project revolutionizes the DNN research by introducing an integrated unsupervised learning computation framework with three vertically-integrated components from the aspects of software (algorithm), hardware (computing), and application (realization). The project echoes the call from the BRAIN Initiative (2013) and the Nanotechnology-Inspired Grand Challenge for Future Computing (2015) from the White House. The research outcomes will benefit both Computational Intelligence (CI) and Computer Architecture (CA) industries at large by introducing a synergy between computing paradigm and artificial intelligence (AI). The corresponding education components enhance existing curricula and pedagogy by introducing interdisciplinary modules on the software/hardware co-design for AI with creative teaching practices, and give special attentions to women and underrepresented minority groups.The project performs three tasks: (1) At the software level, a generalized hierarchical decision-making (GHDM) system is designed to efficiently execute the state-of-the-art unsupervised learning and reinforcement learning processes with substantially reduced computation cost; (2) At the hardware level, a novel DNN computing paradigm is designed with enhanced unsupervised learning supports, based on the novelties in near data computing, GPU architecture, and FGPA + heterogeneous platforms; (3) At the application level, the usage of Ula! is exploited in scenarios that can greatly benefit from unsupervised learning and reinforcement learning. The developed techniques are also demonstrated and evaluated on three representative computing platforms: GPU, FPGA, and emerging nanoscale computing systems, respectively.
期刊论文(32)
专著(0)
科研奖励(0)
会议论文
登录
查看更多内容
DOI:
10.1109/hpca47549.2020.00012
发表时间:
2020-01
期刊:
2020 IEEE International Symposium on High Performance Computer Architecture (HPCA)
影响因子:
--
作者:
[Mingyu Yan;Lei Deng;Xing Hu;Ling Liang;Yujing Feng;Xiaochun Ye;Zhimin Zhang;Dongrui Fan;]
通讯作者:
Mingyu Yan;Lei Deng;Xing Hu;Ling Liang;Yujing Feng;Xiaochun Ye;Zhimin Zhang;Dongrui Fan;
DOI:
10.1609/aaai.v33i01.33011311
发表时间:
2018-09
期刊:
ArXiv
影响因子:
--
作者:
[Yujie Wu;Lei Deng;Guoqi Li;Jun Zhu;Luping Shi]
通讯作者:
Yujie Wu;Lei Deng;Guoqi Li;Jun Zhu;Luping Shi
Rethinking the performance comparison between SNNS and ANNS
重新思考 SNN 和 ANN 之间的性能比较
DOI:
10.1016/j.neunet.2019.09.005
发表时间:
2020-01-01
期刊:
NEURAL NETWORKS
影响因子:
7.8
作者:
[Deng, Lei, Wu, Yujie, Xie, Yuan]
通讯作者:
Xie, Yuan
DOI:
10.1109/isca45697.2020.00082
发表时间:
2020-05
期刊:
2020 ACM/IEEE 47th Annual International Symposium on Computer Architecture (ISCA)
影响因子:
--
作者:
[Yang Zhao;Xiaohan Chen;Yue Wang;Chaojian Li;Haoran You;Y. Fu;Yuan Xie;Zhangyang Wang;Yingyan Lin]
通讯作者:
Yang Zhao;Xiaohan Chen;Yue Wang;Chaojian Li;Haoran You;Y. Fu;Yuan Xie;Zhangyang Wang;Yingyan Lin
NNBench-X: A Benchmarking Methodology for Neural Network Accelerator Designs
NNBench-X:神经网络加速器设计的基准测试方法
DOI:
10.1145/3417709
发表时间:
2020
期刊:
ACM Transactions on Architecture and Code Optimization
影响因子:
1.6
作者:
[Xie, Xinfeng, Hu, Xing, Gu, Peng, Li, Shuangchen, Ji, Yu, Xie, Yuan]
通讯作者:
Xie, Yuan
共 26 条
SHF:SMALL:Collaborative Research: Exploring Nonvolatility of Emerging Memory Technologies for Architecture Design
-
批准号:1816833
-
项目类别:Standard Grant
-
资助金额:$24.9万
-
财政年份:2018
-
负责人:Yuan Xie
-
依托单位:
II-New: RICARDO: Research Infrastructure for Circuit and Architecture Design with Emerging Technologies
-
批准号:1730309
-
项目类别:Standard Grant
-
资助金额:$50.0万
-
财政年份:2017
-
负责人:Yuan Xie
-
依托单位:
XPS: FULL: DSD: Collaborative Research: Parallelizing and Accelerating Metagenomic Applications
-
批准号:1533933
-
项目类别:Standard Grant
-
资助金额:$54.8万
-
财政年份:2015
-
负责人:Yuan Xie
-
依托单位:
SHF: Medium: ASKS - Architecture Support for darK Silicon
-
批准号:1409798
-
项目类别:Standard Grant
-
资助金额:$90.0万
-
财政年份:2014
-
负责人:Yuan Xie
-
依托单位:
SHF: Small: Collaborative Research: STEMS: STatistic Emerging Memory Simulator
-
批准号:1461698
-
项目类别:Standard Grant
-
资助金额:$11.45万
-
财政年份:2014
-
负责人:Yuan Xie
-
依托单位:
SHF: Medium: ASKS - Architecture Support for darK Silicon
-
批准号:1500848
-
项目类别:Standard Grant
-
资助金额:$89.64万
-
财政年份:2014
-
负责人:Yuan Xie
-
依托单位:
SHF: Small: Collaborative Research: STEMS: STatistic Emerging Memory Simulator
-
批准号:1218867
-
项目类别:Standard Grant
-
资助金额:$15.0万
-
财政年份:2012
-
负责人:Yuan Xie
-
依托单位:
ADAMS: Architecture and Design Automation for 3D Multi-core Systems
-
批准号:0903432
-
项目类别:Standard Grant
-
资助金额:$48.0万
-
财政年份:2009
-
负责人:Yuan Xie
-
依托单位:
CSR: Medium: Collaborative Research: Providing Predictable Timing for Task Migration in Embedded Multi-Core Environments (TiME-ME)
-
批准号:0905365
-
项目类别:Continuing Grant
-
资助金额:$33.5万
-
财政年份:2009
-
负责人:Yuan Xie
-
依托单位:
Student Travel Support for International Symposium on High-Performance Computer Architecture (HPCA) 2010
-
批准号:0952841
-
项目类别:Standard Grant
-
资助金额:$4.0万
-
财政年份:2009
-
负责人:Yuan Xie
-
依托单位:
CAREER: Process Variation Aware Embedded MPSoC Synthesis
-
批准号:0643902
-
项目类别:Continuing Grant
-
资助金额:$40.0万
-
财政年份:2007
-
负责人:Yuan Xie
-
依托单位:
CSR--EHS: Collaborative Research: Hybrid Timing Analysis via Multi-Mode Execution
-
批准号:0720659
-
项目类别:Standard Grant
-
资助金额:$0.0万
-
财政年份:2007
-
负责人:Yuan Xie
-
依托单位:
海外基金