arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1910.09092cs.LGmath.OCstat.MEstat.ML
收藏

快速精确矩阵完备化:一个统一的矩阵完备化优化框架

Fast Exact Matrix Completion: A Unified Optimization Framework for Matrix Completion

Sloan School of Management(斯隆管理学院);Operations Research Center(运筹学研究中心);Massachusetts Institute of Technology(麻省理工学院)

Dimitris Bertsimas, Michael Lingzhi Li

摘要:

我们制定的矩阵完成的问题,没有边信息作为一个非凸优化问题。我们设计的fastImpute基于非凸梯度下降,并显示它收敛到一个全局最小值,保证恢复密切的基础矩阵,而它的规模超过$10^5 \times 10^5$的大小矩阵。我们报告了在合成和真实世界数据集上的实验,这些实验表明fastImpute在恢复矩阵的准确性和所有情况下所需的时间方面都具有竞争力。此外,当大量条目丢失时,fastImpute在具有和不具有边信息的情况下都比当前最先进的矩阵完成方法快$75\%$以上、低MAPE和快$15$倍。

英文摘要:

We formulate the problem of matrix completion with and without side information as a non-convex optimization problem. We design fastImpute based on non-convex gradient descent and show it converges to a global minimum that is guaranteed to recover closely the underlying matrix while it scales to matrices of sizes beyond $10^5 \times 10^5$. We report experiments on both synthetic and real-world datasets that show fastImpute is competitive in both the accuracy of the matrix recovered and the time needed across all cases. Furthermore, when a high number of entries are missing, fastImpute is over $75\%$ lower in MAPE and $15$ times faster than current state-of-the-art matrix completion methods in both the case with side information and without.

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。

↑