Curriculum Learning for Handwritten Text Line Recognition

TL;DR

This paper introduces a curriculum learning approach to improve the training efficiency and performance of RNNs in handwritten text line recognition by gradually increasing sequence length during training.

Contribution

It proposes a simple curriculum learning method for sequence recognition tasks, accelerating training and enhancing accuracy in handwritten text recognition.

Findings

Training time reduced significantly with curriculum learning.

Performance improved on multiple handwritten text datasets.

Simple implementation makes it practical for real-world use.

Abstract

Recurrent Neural Networks (RNN) have recently achieved the best performance in off-line Handwriting Text Recognition. At the same time, learning RNN by gradient descent leads to slow convergence, and training times are particularly long when the training database consists of full lines of text. In this paper, we propose an easy way to accelerate stochastic gradient descent in this set-up, and in the general context of learning to recognize sequences. The principle is called Curriculum Learning, or shaping. The idea is to first learn to recognize short sequences before training on all available training sequences. Experiments on three different handwritten text databases (Rimes, IAM, OpenHaRT) show that a simple implementation of this strategy can significantly speed up the training of RNN for Text Recognition, and even significantly improve performance in some cases.