CRESPR: Modular sparsification of DNNs to improve pruning performance and model interpretability

Tianyu Kang; Wei Ding; Ping Chen

doi:10.1016/j.neunet.2023.12.021

CRESPR: Modular sparsification of DNNs to improve pruning performance and model interpretability

Tianyu Kang
, Wei Ding
, Ping Chen

University of Massachusetts Boston

Research output: Contribution to journal › Article › peer-review

Original language	English
Article number	106067
Journal	Neural Networks
Volume	172
DOIs	https://doi.org/10.1016/j.neunet.2023.12.021
State	Published - Apr 2024

ASJC Scopus Subject Areas

Cognitive Neuroscience
Artificial Intelligence

Keywords

Model interpretability
Neural network pruning

Access to Document

10.1016/j.neunet.2023.12.021

Cite this

@article{a2d7a2dc72184bbca7d5dd2cd92a3538,

title = "CRESPR: Modular sparsification of DNNs to improve pruning performance and model interpretability",

keywords = "Model interpretability, Neural network pruning",

author = "Tianyu Kang and Wei Ding and Ping Chen",

note = "Publisher Copyright: {\textcopyright} 2023 Elsevier Ltd",

year = "2024",

month = apr,

doi = "10.1016/j.neunet.2023.12.021",

language = "English",

volume = "172",

journal = "Neural Networks",

issn = "0893-6080",

publisher = "Elsevier Ltd",

}

TY - JOUR

T1 - CRESPR

T2 - Modular sparsification of DNNs to improve pruning performance and model interpretability

AU - Kang, Tianyu

AU - Ding, Wei

AU - Chen, Ping

PY - 2024/4

Y1 - 2024/4

KW - Model interpretability

KW - Neural network pruning

UR - https://www.scopus.com/pages/publications/85181957660

UR - https://www.scopus.com/pages/publications/85181957660#tab=citedBy

U2 - 10.1016/j.neunet.2023.12.021

DO - 10.1016/j.neunet.2023.12.021

M3 - Article

C2 - 38199151

AN - SCOPUS:85181957660

SN - 0893-6080

VL - 172

JO - Neural Networks

JF - Neural Networks

M1 - 106067

ER -

CRESPR: Modular sparsification of DNNs to improve pruning performance and model interpretability

ASJC Scopus Subject Areas

Keywords

Access to Document

Other files and links

Fingerprint

Cite this