arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.04883cs.LGstat.ML
收藏

半监督向量值学习:改进的界与算法

Semi-supervised Vector-valued Learning: Improved Bounds and Algorithms

Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);Gaoling School of Artificial Intelligence, Renmin University of China(中国人民大学高瓴人工智能学院)

Jian Li, Yong Liu, Weiping Wang

AI总结:

针对向量值学习问题,结合局部Rademacher复杂度与未标记数据推导更紧的半监督超额风险界,据此提出融合拉普拉斯正则化的通用算法,实验验证其性能更优。

摘要:

向量值学习(输出空间具有向量值结构)是一个重要问题,涵盖了广泛的重要领域,例如多任务学习和迁移学习。利用局部Rademacher复杂度与未标记数据,我们从核视角和线性视角出发,为一般向量值学习推导了新颖的半监督超额风险界。所推导的界比现有界更紧,收敛速率从标记样本量的平方根改进为总样本量的平方根或直接依赖于标记样本量。受理论分析的启发,我们提出了一种通用的半监督算法,用于高效学习向量值函数,该算法结合了局部Rademacher复杂度与拉普拉斯正则化。大量实验结果表明,所提出的算法显著优于对比方法,这与我们的理论发现一致。

英文摘要:

Vector-valued learning, where the output space admits a vector-valued structure, is an important problem 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 regularization. Extensive experimental results illustrate the proposed algorithm significantly outperforms the compared methods, which coincides with our theoretical findings.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑