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
Damping proximal coordinate descent algorithm for non-convex regularization
TL;DR: The Damping Proximal Coordinate Descent algorithms are proposed that address the optimization issues of a general family of non-convex regularized problems and are guaranteed to be globally convergent.
Posted Content
CATRO: Channel Pruning via Class-Aware Trace Ratio Optimization.
TL;DR: In this article, the authors proposed a channel pruning method via class-aware trace ratio optimization (CATRO), which measures the joint impact of channels by feature space discriminations and consolidates the layer-wise impact of preserved channels.
Journal ArticleDOI
Bipartite Ranking Fairness through a Model Agnostic Ordering Adjustment
TL;DR: In this article , a model agnostic post-processing framework is proposed to learn a ranking function that ranks positive instances higher than negative ones for bipartite ranking and maintaining the algorithm classification performance.
Proceedings ArticleDOI
Image Captioning with Partially Rewarded Imitation Learning
TL;DR: Experiments demonstrate that the proposed novel training objective for image captioning can integrate the strengths of state-of-the-arts, producing more human-like captions while maintaining comparable performance on traditional metrics.
Proceedings ArticleDOI
Instance- and bag-level manifold regularization for aggregate outputs classification
TL;DR: A manifold regularization framework is set up to deal with the aggregate outputs classification problem, and four concrete algorithms are proposed, each of which can cope with both binary and multi-class scenarios.