arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1703.02492cs.LGstat.ML
收藏

在线多线性词典学习

Online Multilinear Dictionary Learning

Department of Electrical & Electronic Engineering, Imperial College London, United Kingdom(伦敦帝国理工学院电子与电气工程系)

Thiernithi Variddhisai, Danilo Mandic

摘要:

提出了一种在线张量字典学习方法。在可分字典的假设下,利用张量收缩将$\mathcal{O}\left(L^N\right)$的$N$ -路模型简化为具有实时性的简单的$\mathcal{O}\left(NL^2\right)$矩阵方程。为了避免稀疏矩阵求逆引起的数值不稳定性,通过最小二乘解构造了一类具有记忆的随机梯度算法,以保证算法的收敛性和鲁棒性。讨论并实现了精确线搜索的梯度下降法和牛顿法。还详细解释了如何处理不良初始化和离群值的扩展。两个合成信号的实验证实了我们所提出的方法令人印象深刻的性能。

英文摘要:

A method for online tensor dictionary learning is proposed. With the assumption of separable dictionaries, tensor contraction is used to diminish a $N$-way model of $\mathcal{O}\left(L^N\right)$ into a simple matrix equation of $\mathcal{O}\left(NL^2\right)$ with a real-time capability. To avoid numerical instability due to inversion of sparse matrix, a class of stochastic gradient with memory is formulated via a least-square solution to guarantee convergence and robustness. Both gradient descent with exact line search and Newton's method are discussed and realized. Extensions onto how to deal with bad initialization and outliers are also explained in detail. Experiments on two synthetic signals confirms an impressive performance of our proposed method.

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。

↑