Zobrazeno 1 - 10
of 204
pro vyhledávání: '"Li, Qianyi"'
Continual learning (CL) enables animals to learn new tasks without erasing prior knowledge. CL in artificial neural networks (NNs) is challenging due to catastrophic forgetting, where new learning degrades performance on older tasks. While various te
Externí odkaz:
http://arxiv.org/abs/2407.10315
Artificial neural networks have revolutionized machine learning in recent years, but a complete theoretical framework for their learning process is still lacking. Substantial progress has been made for infinitely wide networks. In this regime, two di
Externí odkaz:
http://arxiv.org/abs/2309.04522
Autor:
Li, Qianyi1,2,3,4 (AUTHOR), Gao, Shuaiyun2,3,4 (AUTHOR), Qi, Yang2 (AUTHOR), Shi, Nuo5 (AUTHOR), Wang, Zhenzhen5 (AUTHOR), Saiding, Qimanguli1 (AUTHOR), Chen, Liang1 (AUTHOR), Du, Yawei1 (AUTHOR), Wang, Bo3,4 (AUTHOR), Yao, Wenfei2 (AUTHOR), Sarmento, Bruno6,7 (AUTHOR), Yu, Jie2 (AUTHOR), Lu, Yiming2,3,4,8 (AUTHOR) luyiming@rjh.com.cn, Wang, Juan1 (AUTHOR) wj12496@rjh.com.cn, Cui, Wenguo1 (AUTHOR) wgcui@sjtu.edu.cn
Publikováno v:
Advanced Science. 11/6/2024, Vol. 11 Issue 41, p1-19. 19p.
Autor:
Li, Qianyi, Sompolinsky, Haim
Publikováno v:
NeurIPS 2022
Recently proposed Gated Linear Networks present a tractable nonlinear network architecture, and exhibit interesting capabilities such as learning with local error signals and reduced forgetting in sequential learning. In this work, we introduce a nov
Externí odkaz:
http://arxiv.org/abs/2210.17449
Autor:
Wang, Yifei, Li, Qianyi, Li, Haoyang, He, Linyan, Xu, Linji, Li, Lin, Wang, Ruizhi, Zhao, Xueyu, Chen, Yongdong, Gu, Li, Li, Jinze, He, Qiang
Publikováno v:
In Renewable Energy December 2024 237 Part C
Autor:
Li, Qianyi1 (AUTHOR), Wu, Chunxuan1 (AUTHOR), Sun, Shiqun1 (AUTHOR), Yang, Lingchao1 (AUTHOR), Li, Yanyan1 (AUTHOR), Niu, Yixin2 (AUTHOR), Zhang, Li1 (AUTHOR), Li, Wei1 (AUTHOR), Yu, Ying1 (AUTHOR) yuying@xinhuamed.com.cn
Publikováno v:
Journal of Diabetes. Jul2024, Vol. 16 Issue 7, p1-10. 10p.
Publikováno v:
Waike lilun yu shijian, Vol 28, Iss 06, Pp 540-550 (2023)
Objective To assess the comparative efficacy of different methods for difficult biliary cannulation in endoscopic retrograde cholangio-pancreatography (ERCP) through a network meta-analysis. Methods Randomized controlled trials (RCTs) that compared t
Externí odkaz:
https://doaj.org/article/39846bade72c4e9eaac093a12a0ffcae
Autor:
Wang, Zhenzhen, Yuan, Jumao, Xu, Yan, Shi, Nuo, Lin, Lin, Wang, Ruirui, Dai, Rong, Xu, Lin, Hao, Ning, Li, Qianyi
Publikováno v:
In Materials Today Bio June 2024 26
Autor:
Azmat, Mian, Yang, Junlin, Li, Qianyi, Zhang, Jingyao, Haibo, Jin, Muhammad Kashif, Naseem, Li, Jingbo
Publikováno v:
In Ceramics International 1 April 2024 50(7) Part A:11119-11128
Autor:
Li, Qianyi, Sompolinsky, Haim
Publikováno v:
Phys. Rev. X 11, 031059 (2021)
The success of deep learning in many real-world tasks has triggered an intense effort to understand the power and limitations of deep learning in the training and generalization of complex tasks, so far with limited progress. In this work, we study t
Externí odkaz:
http://arxiv.org/abs/2012.04030