Zobrazeno 1 - 10
of 4 946
pro vyhledávání: '"GONG, Yu"'
Fast-converging algorithms are a contemporary requirement in reinforcement learning. In the context of linear function approximation, the magnitude of the smallest eigenvalue of the key matrix is a major factor reflecting the convergence speed. Tradi
Externí odkaz:
http://arxiv.org/abs/2411.06396
Autor:
Yang, Cheng, Sui, Yang, Xiao, Jinqi, Huang, Lingyi, Gong, Yu, Duan, Yuanlin, Jia, Wenqi, Yin, Miao, Cheng, Yu, Yuan, Bo
The emergence of Mixture of Experts (MoE) LLMs has significantly advanced the development of language models. Compared to traditional LLMs, MoE LLMs outperform traditional LLMs by achieving higher performance with considerably fewer activated paramet
Externí odkaz:
http://arxiv.org/abs/2411.01016
Low-rank compression, a popular model compression technique that produces compact convolutional neural networks (CNNs) with low rankness, has been well-studied in the literature. On the other hand, low-rank training, as an alternative way to train lo
Externí odkaz:
http://arxiv.org/abs/2401.10341
Publikováno v:
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:38125-38136, 2023
Attention-based vision models, such as Vision Transformer (ViT) and its variants, have shown promising performance in various computer vision tasks. However, these emerging architectures suffer from large model sizes and high computational costs, cal
Externí odkaz:
http://arxiv.org/abs/2305.17235
Publikováno v:
Zhejiang dianli, Vol 43, Iss 10, Pp 35-44 (2024)
As the penetration rate of renewable energy sources increases in new-type power systems， so too does the complexity of the grid structure， leading to more diverse and complex power quality disturbance （PQD）. To accurately identify power quali
Externí odkaz:
https://doaj.org/article/0c46d0609b154622917f70d4e3ee1e2a
Publikováno v:
ACS Omega, Vol 9, Iss 42, Pp 43106-43114 (2024)
Externí odkaz:
https://doaj.org/article/b0a6274c6c294056a90abb485874792b
Publikováno v:
生物医学转化, Vol 5, Iss 3, Pp 55-60 (2024)
Objective To compare the clinical efficacy of arthroscopic closed reduction internal fixation with Open Reduction Internal Fixation (ORIF) for medial malleolus fractures. Method A total of 43 patients with medial malleolus fractures treated at Jinjia
Externí odkaz:
https://doaj.org/article/6fd3c2e3890a4089b0f423928d157d10
Visual-based defect detection is a crucial but challenging task in industrial quality control. Most mainstream methods rely on large amounts of existing or related domain data as auxiliary information. However, in actual industrial production, there
Externí odkaz:
http://arxiv.org/abs/2304.13963
Publikováno v:
Industrial Management & Data Systems, 2024, Vol. 124, Issue 8, pp. 2633-2662.
Externí odkaz:
http://www.emeraldinsight.com/doi/10.1108/IMDS-11-2023-0877
Autor:
Xiao, Jinqi, Zhang, Chengming, Gong, Yu, Yin, Miao, Sui, Yang, Xiang, Lizhi, Tao, Dingwen, Yuan, Bo
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence. 37, 9 (Jun. 2023), 10464-10472
Low-rank compression is an important model compression strategy for obtaining compact neural network models. In general, because the rank values directly determine the model complexity and model accuracy, proper selection of layer-wise rank is very c
Externí odkaz:
http://arxiv.org/abs/2301.09422