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 1808.03351stat.MLcs.LG
收藏

利用结构实现快速核学习

Exploiting Structure for Fast Kernel Learning

University of Toronto(多伦多大学)

Trefor W. Evans, Prasanth B. Nair

摘要:

我们提出了两种方法,用于精确的高斯过程(GP)推理和学习大量的图像,视频,时空,或多输出数据集的缺失值(或“间隙”)在观察到的响应。第一种方法利用稀疏选择矩阵忽略间隙,并引入一个高效的低秩预条件子来加速计算。第二种方法引入了一种新的GP训练方法,即在显式训练模型之前根据间隙推断响应值。我们发现这第二种方法是非常有利的一类问题考虑。这两种新的方法都广泛使用Kronecker矩阵代数来设计具有低内存需求的大规模可扩展算法。我们展示了精确的GP推理的时空气候建模问题与370万个训练点,以及视频重建问题与10亿点。

英文摘要:

We propose two methods for exact Gaussian process (GP) inference and learning on massive image, video, spatial-temporal, or multi-output datasets with missing values (or "gaps") in the observed responses. The first method ignores the gaps using sparse selection matrices and a highly effective low-rank preconditioner is introduced to accelerate computations. The second method introduces a novel approach to GP training whereby response values are inferred on the gaps before explicitly training the model. We find this second approach to be greatly advantageous for the class of problems considered. Both of these novel approaches make extensive use of Kronecker matrix algebra to design massively scalable algorithms which have low memory requirements. We demonstrate exact GP inference for a spatial-temporal climate modelling problem with 3.7 million training points as well as a video reconstruction problem with 1 billion points.

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。

↑