Zobrazeno 1 - 10
of 17
pro vyhledávání: '"Shangling Jui"'
Autor:
Keith G. Mills, Mohammad Salameh, Di Niu, Fred X. Han, Seyed Saeed Changiz Rezaei, Hengshuai Yao, Wei Lu, Shuo Lian, Shangling Jui
Publikováno v:
IEEE Access, Vol 9, Pp 110962-110974 (2021)
Recent developments in Neural Architecture Search (NAS) resort to training the supernet of a predefined search space with weight sharing to speed up architecture evaluation. These include random search schemes, as well as various schemes based on opt
Externí odkaz:
https://doaj.org/article/a1e28986bd9d431aa6d51fab7a428655
Publikováno v:
Proceedings of the International Symposium on Combinatorial Search. 15:74-82
Beam search is a popular algorithm for solving real-world problems --- especially where search space is an enormously large tree but real-time solutions are most preferred. We present a memory-bounded best-first beam search (MB2FBS), which can be vie
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ae7a459db11ac2c65d55e8c2c67c0f45
https://doi.org/10.1609/socs.v15i1.21754
https://doi.org/10.1609/socs.v15i1.21754
Autor:
Fred X. Han, Keith G. Mills, Fabian Chudak, Parsa Riahi, Mohammad Salameh, Jialin Zhang, Wei Lu, Shangling Jui, Di Niu
Publikováno v:
Proceedings of the 2023 SIAM International Conference on Data Mining (SDM) ISBN: 9781611977653
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::688d3ae01cb06ec4e3a6e3b1297bf33b
https://doi.org/10.1137/1.9781611977653.ch81
https://doi.org/10.1137/1.9781611977653.ch81
Most meta-learning approaches assume the existence of a very large set of labeled data available for episodic meta-learning of base knowledge. This contrasts with the more realistic continual learning paradigm in which data arrives incrementally in t
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7140746e44412920aecf158286c83317
http://arxiv.org/abs/2111.04993
http://arxiv.org/abs/2111.04993
Autor:
Chao Gao, Tong Mo, Taylor Zowtuk, Tanvir Sajed, Laiyuan Gong, Hanxuan Chen, Shangling Jui, Wei Lu
Publikováno v:
2021 IEEE 33rd International Conference on Tools with Artificial Intelligence (ICTAI).
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::f3f7faf9f02e08d50f4740834bcb3420
https://doi.org/10.1109/ictai52525.2021.00045
https://doi.org/10.1109/ictai52525.2021.00045
Autor:
Seyed Saeed Changiz Rezaei, Wei Lu, Di Niu, Fred X. Han, Jialin Zhang, Fabian Chudak, Shuo Lian, Shangling Jui, Keith G. Mills
Publikováno v:
CIKM
Neural architecture search automates neural network design and has achieved state-of-the-art results in many deep learning applications. While recent literature has focused on designing networks to maximize accuracy, little work has been conducted to
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::641d909ce238bcda78b0357c67018dc0
https://doi.org/10.1145/3459637.3481944
https://doi.org/10.1145/3459637.3481944
Publikováno v:
Proceedings of the Future Technologies Conference (FTC) 2021, Volume 1 ISBN: 9783030899059
Many interesting problems in statistics and machine learning can be written as \(min_xF(x)=f(x)+g(x)\), where \(x\) is the model parameter, \(f\) is the loss and \(g\) is the regularizer. Examples include regularized regression in high-dimensional fe
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f37fbccba0627709d8947d255c9b9114
https://kar.kent.ac.uk/78761/1/FTC_2021_Nonsmooth.pdf
https://kar.kent.ac.uk/78761/1/FTC_2021_Nonsmooth.pdf
Autor:
Min Qin, Cheng Chen, Ruosheng Xu, Xin Huang, Shangling Jui, Pengyun Li, Zhihao Ding, Yu Huang
Publikováno v:
ITC
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::69a0046aac1b7c7ed668a4ab3d5ea1b0
https://doi.org/10.1109/itc50571.2021.00010
https://doi.org/10.1109/itc50571.2021.00010
Domain adaptation (DA) aims to transfer the knowledge learned from a source domain to an unlabeled target domain. Some recent works tackle source-free domain adaptation (SFDA) where only a source pre-trained model is available for adaptation to the t
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::02b4ad82645db469f2ab1ee06fe61d58
http://arxiv.org/abs/2108.01614
http://arxiv.org/abs/2108.01614
Publikováno v:
CVPR
An effective and efficient architecture performance evaluation scheme is essential for the success of Neural Architecture Search (NAS). To save computational cost, most of existing NAS algorithms often train and evaluate intermediate neural architect
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::80e4b8ebc68171a06f1ba671c75f86e1
https://doi.org/10.1109/cvpr46437.2021.00439
https://doi.org/10.1109/cvpr46437.2021.00439