.png){kind=icon}
返回
Semi-supervised vector-valued learning: Improved bounds and algorithms
.png){kind=icon}
OA
AI
DOI：10.1016/j.patcog.2023.109356.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Vector-valued learning, where the output space admits a vector-valued structure, is an important prob-lem that covers a broad family of important domains, e.g. multi-task learning and transfer learning. Using local Rademacher complexity and unlabeled data, we derive novel semi-supervised excess risk bounds for general vector-valued learning from both kernel perspective and linear perspective. The derived bounds are much sharper than existing ones and the convergence rates are improved from the square root of labeled sample size to the square root of total sample size or directly dependent on labeled sample size. Motivated by our theoretical analysis, we propose a general semi-supervised algorithm for efficiently learning vector-valued functions, incorporating both local Rademacher complexity and Laplacian regu-larization. Extensive experimental results illustrate the proposed algorithm significantly outperforms the compared methods, which coincides with our theoretical findings.(c) 2023 Elsevier Ltd. All rights reserved.
Keyword:
Vector-valued learning
Semi-supervised learning
Excess risk bound
Local rademacher complexity
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
7.6
论文数:
1.3W
被引数:
4.5W
机构
引用论文
RANDOMIZED SKETCHES FOR KERNELS: FAST AND OPTIMAL NONPARAMETRIC REGRESSION内核的随机草图: 快速和最佳的非参数回归.png)
ANNALS OF STATISTICSOAAI
ANNALS OF STATISTICSIF3.7
OAAI分享收藏Online Adaptive Kernel Learning with Random Features for Large-scale Nonlinear ClassificationPATTERN RECOGNITION
PATTERN RECOGNITIONIF7.6
PREAI
分享收藏分享收藏
查看更多内容