Regular expressions for decoding of neural network outputs

TL;DR

This paper introduces a regular expression-based decoding method for neural network outputs in handwritten text recognition, offering efficiency and applicability across various tasks.

Contribution

It presents a novel, efficient decoding approach using regular expressions and finite automata for CTC-trained neural networks, with theoretical analysis and practical approximation strategies.

Findings

Significant speed-up over existing decoding methods

Effective in diverse applications from information retrieval to full text recognition

Approximation method remains reliable when regular expression matches ground truth

Abstract

This article proposes a convenient tool for decoding the output of neural networks trained by Connectionist Temporal Classification (CTC) for handwritten text recognition. We use regular expressions to describe the complex structures expected in the writing. The corresponding finite automata are employed to build a decoder. We analyze theoretically which calculations are relevant and which can be avoided. A great speed-up results from an approximation. We conclude that the approximation most likely fails if the regular expression does not match the ground truth which is not harmful for many applications since the low probability will be even underestimated. The proposed decoder is very efficient compared to other decoding methods. The variety of applications reaches from information retrieval to full text recognition. We refer to applications where we integrated the proposed decoder successfully.