Výsledky vyhledávání - "Jiang Chonghe"

Report

GETS: Ensemble Temperature Scaling for Calibration in Graph Neural Networks

Autor: Zhuang, Dingyi, Jiang, Chonghe, Zheng, Yunhan, Wang, Shenhao, Zhao, Jinhua

Graph Neural Networks deliver strong classification results but often suffer from poor calibration performance, leading to overconfidence or underconfidence. This is particularly problematic in high stakes applications where accurate uncertainty esti

Externí odkaz: http://arxiv.org/abs/2410.09570

Zobrazit plný text záznamu

Report

Computing Competitive Equilibrium for Chores: Linear Convergence and Lightweight Iteration

Autor: Chen, He, Jiang, Chonghe, So, Anthony Man-Cho

Competitive equilibrium (CE) for chores has recently attracted significant attention, with many algorithms proposed to approximately compute it. However, existing algorithms either lack iterate convergence guarantees to an exact CE or require solving

Externí odkaz: http://arxiv.org/abs/2410.04036

Zobrazit plný text záznamu

Report

Graph Sparsification via Mixture of Graphs

Autor: Zhang, Guibin, Sun, Xiangguo, Yue, Yanwei, Jiang, Chonghe, Wang, Kun, Chen, Tianlong, Pan, Shirui

Graph Neural Networks (GNNs) have demonstrated superior performance across various graph learning tasks but face significant computational challenges when applied to large-scale graphs. One effective approach to mitigate these challenges is graph spa

Externí odkaz: http://arxiv.org/abs/2405.14260

Zobrazit plný text záznamu

Report

On the Estimation Performance of Generalized Power Method for Heteroscedastic Probabilistic PCA

Autor: Wang, Jinxin, Jiang, Chonghe, Liu, Huikang, So, Anthony Man-Cho

The heteroscedastic probabilistic principal component analysis (PCA) technique, a variant of the classic PCA that considers data heterogeneity, is receiving more and more attention in the data science and signal processing communities. In this paper,

Externí odkaz: http://arxiv.org/abs/2312.03438

Zobrazit plný text záznamu

Report

Foundation Model is Efficient Multimodal Multitask Model Selector

Autor: Meng, Fanqing, Shao, Wenqi, Peng, Zhanglin, Jiang, Chonghe, Zhang, Kaipeng, Qiao, Yu, Luo, Ping

This paper investigates an under-explored but important problem: given a collection of pre-trained neural networks, predicting their performance on each multi-modal task without fine-tuning them, such as image recognition, referring, captioning, visu

Externí odkaz: http://arxiv.org/abs/2308.06262

Zobrazit plný text záznamu