Jiyong Yu

Dr. Jiyong Yu

orcid

Research Focus & Metrics

Academic Career Length

21Years (2006-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 April 2020 in IEEE Micro
Verified
Journal Article
Closed access
JYJiyong Yu
Jiyong Yu
MYMengjia Yan
Mengjia Yan
AKArtem Khyzha
Artem Khyzha
AMAdam Morrison
Adam Morrison
JTJosep Torrellas
Josep Torrellas
CFChristopher W. Fletcher
Christopher W. Fletcher

27 February 2006 in Applied Physics Letters
Verified
Journal Article
Open access
PDF file
HZHongzhou Zhang
Hongzhou Zhang
MPM. R. Phillips
M. R. Phillips
JGJ. D. Fitz Gerald
J. D. Fitz Gerald
JYJiyong Yu
Jiyong Yu
Ying (Ian) ChenYing (Ian) Chen
Ying (Ian) Chen
We demonstrate a simple and effective approach for growing large-scale, high-density, and well-patterned conical boron nitride nanorods. A catalyst layer of Fe(NO3)3 was patterned on a silicon substrate by using a copper grid as a mask. The nanorods were grown via annealing milled boron carbide powders at 1300°C in a flow of nitrogen gas. The as-grown nanorods exhibit uniform morphology and the catalyst pattern precisely defines the position of nanorod deposition. Cathodoluminescence (CL) spectra of the nanorods show two broad emission bands centered at 3.75 and 1.85eV . Panchromatic CL images reveal clear patterned structure.


6 April 2020 in IEEE Micro
Verified
Journal Article
Closed access
JYJiyong Yu
Jiyong Yu
LHLucas Hsiung
Lucas Hsiung
MHMohamad El Hajj
Mohamad El Hajj
CFChristopher W. Fletcher
Christopher W. Fletcher

18 April 2018 in arXiv
Verified
Preprint
Open access
PDF file
KHKartik Hegde
Kartik Hegde
JYJiyong Yu
Jiyong Yu
RARohit Agrawal
Rohit Agrawal
MYMengjia Yan
Mengjia Yan
MPMichael Pellauer
Michael Pellauer
CFChristopher W. Fletcher
Christopher W. Fletcher
Convolutional Neural Networks (CNNs) have begun to permeate all corners of electronic society (from voice recognition to scene generation) due to their high accuracy and machine efficiency per operation. At their core, CNN computations are made up of multi-dimensional dot products between weight and input vectors. This paper studies how weight repetition ---when the same weight occurs multiple times in or across weight vectors--- can be exploited to save energy and improve performance during CNN inference. This generalizes a popular line of work to improve efficiency from CNN weight sparsity, as reducing computation due to repeated zero weights is a special case of reducing computation due to repeated weights. To exploit weight repetition, this paper proposes a new CNN accelerator called the Unique Weight CNN Accelerator (UCNN). UCNN uses weight repetition to reuse CNN sub-computations (e.g., dot products) and to reduce CNN model size when stored in off-chip DRAM ---both of which save energy. UCNN further improves performance by exploiting sparsity in weights. We evaluate UCNN with an accelerator-level cycle and energy model and with an RTL implementation of the UCNN processing element. On three contemporary CNNs, UCNN improves throughput-normalized energy consumption by 1.2x - 4x, relative to a similarly provisioned baseline accelerator that uses Eyeriss-style sparsity optimizations. At the same time, the UCNN processing element adds only 17-24% area overhead relative to the same baseline.