Adaptive Kernel Value Caching for SVM Training

TL;DR

This paper introduces adaptive kernel value caching strategies, EFU and HCST, that significantly improve SVM training efficiency by increasing cache hit ratios and reducing training time, especially on large, high-dimensional datasets.

Contribution

The paper proposes novel adaptive caching strategies, EFU and HCST, that outperform traditional LRU caching in SVM training by dynamically adjusting to training stages.

Findings

EFU achieves 20% higher hit ratio than LRU with Gaussian kernels.

HCST adaptively maintains high hit ratios across training stages.

HCST reduces training time by 20% on average compared to existing strategies.

Abstract

Support Vector Machines (SVMs) can solve structured multi-output learning problems such as multi-label classification, multiclass classification and vector regression. SVM training is expensive especially for large and high dimensional datasets. The bottleneck of the SVM training often lies in the kernel value computation. In many real-world problems, the same kernel values are used in many iterations during the training, which makes the caching of kernel values potentially useful. The majority of the existing studies simply adopt the LRU (least recently used) replacement strategy for caching kernel values. However, as we analyze in this paper, the LRU strategy generally achieves high hit ratio near the final stage of the training, but does not work well in the whole training process. Therefore, we propose a new caching strategy called EFU (less frequently used) which replaces the less frequently used kernel values that enhances LFU (least frequently used). Our experimental results show that EFU often has 20\% higher hit ratio than LRU in the training with the Gaussian kernel. To further optimize the strategy, we propose a caching strategy called HCST (hybrid caching for the SVM training), which has a novel mechanism to automatically adapt the better caching strategy in the different stages of the training. We have integrated the caching strategy into ThunderSVM, a recent SVM library on many-core processors. Our experiments show that HCST adaptively achieves high hit ratios with little runtime overhead among different problems including multi-label classification, multiclass classification and regression problems. Compared with other existing caching strategies, HCST achieves 20\% more reduction in training time on average.