# Squeezed Very Deep Convolutional Neural Networks for Text Classification

## TL;DR

This paper introduces SVDCNN, a compact and efficient deep convolutional neural network for text classification, optimized for mobile devices with minimal accuracy loss and reduced latency.

## Contribution

The paper proposes a modified VDCNN architecture using depthwise separable convolutions and global pooling to significantly reduce size and latency while maintaining accuracy.

## Key findings

- Model size reduced by 10x to 20x
- Achieves up to 6MB size
- Loss in accuracy between 0.4% and 1.3%

## Abstract

Most of the research in convolutional neural networks has focused on increasing network depth to improve accuracy, resulting in a massive number of parameters which restricts the trained network to platforms with memory and processing constraints. We propose to modify the structure of the Very Deep Convolutional Neural Networks (VDCNN) model to fit mobile platforms constraints and keep performance. In this paper, we evaluate the impact of Temporal Depthwise Separable Convolutions and Global Average Pooling in the network parameters, storage size, and latency. The squeezed model (SVDCNN) is between 10x and 20x smaller, depending on the network depth, maintaining a maximum size of 6MB. Regarding accuracy, the network experiences a loss between 0.4% and 1.3% and obtains lower latencies compared to the baseline model.