Training with reduced precision of a support vector machine model for text classification
This work addresses efficiency improvements for text classification models on hardware platforms like GPUs and FPGAs, but appears incremental as it applies existing quantization techniques to SVMs.
The paper investigated how quantization affects the efficiency of training a support vector machine (SVM) for multi-class text classification, finding that reduced precision decreased computation time and memory usage while maintaining accuracy.
This paper presents the impact of using quantization on the efficiency of multi-class text classification in the training process of a support vector machine (SVM). This work is focused on comparing the efficiency of SVM model trained using reduced precision with its original form. The main advantage of using quantization is decrease in computation time and in memory footprint on the dedicated hardware platform which supports low precision computation like GPU (16-bit) or FPGA (any bit-width). The paper presents the impact of a precision reduction of the SVM training process on text classification accuracy. The implementation of the CPU was performed using the OpenMP library. Additionally, the results of the implementation of the GPU using double, single and half precision are presented.