C
Changshui Zhang
Researcher at Tsinghua University
Publications - 509
Citations - 22100
Changshui Zhang is an academic researcher from Tsinghua University. The author has contributed to research in topics: Semi-supervised learning & Support vector machine. The author has an hindex of 67, co-authored 493 publications receiving 18471 citations. Previous affiliations of Changshui Zhang include Microsoft & Cornell University.
Papers
More filters
Journal ArticleDOI
Spectral Embedded Clustering: A Framework for In-Sample and Out-of-Sample Spectral Clustering
TL;DR: This paper proposes the spectral embedded clustering (SEC) framework, in which a linearity regularization is explicitly added into the objective function of SC methods, and presents a new Laplacian matrix constructed from a local regression of each pattern to capture both local and global discriminative information for clustering.
Proceedings ArticleDOI
Robust multi-task feature learning
TL;DR: This paper proposes a Robust Multi-Task Feature Learning algorithm (rMTFL) which simultaneously captures a common set of features among relevant tasks and identifies outlier tasks, and provides a detailed theoretical analysis on the proposed rMTFL formulation.
Journal ArticleDOI
Trace Ratio Problem Revisited
TL;DR: A theoretical overview of the global optimum solution to the TR problem via the equivalent trace difference problem is proposed, and Eigenvalue perturbation theory is introduced to derive an efficient algorithm based on the Newton-Raphson method.
Proceedings ArticleDOI
Recurrent Convolutional Neural Networks for Continuous Sign Language Recognition by Staged Optimization
TL;DR: This work presents a weakly supervised framework with deep neural networks for vision-based continuous sign language recognition, where the ordered gloss labels but no exact temporal locations are available with the video of sign sentence, and the amount of labeled sentences for training is limited.
Journal ArticleDOI
An in-field automatic wheat disease diagnosis system
TL;DR: Experimental results demonstrate that the proposed system outperforms conventional CNN architectures on recognition accuracy under the same amount of parameters, meanwhile maintaining accurate localization for corresponding disease areas.