Timezone: »

 
Poster
Connectionist Temporal Classification with Maximum Entropy Regularization
Hu Liu · Sheng Jin · Changshui Zhang

Thu Dec 06 07:45 AM -- 09:45 AM (PST) @ Room 517 AB #106
in Thu Poster Session A »

Connectionist Temporal Classification (CTC) is an objective function for end-to-end sequence learning, which adopts dynamic programming algorithms to directly learn the mapping between sequences. CTC has shown promising results in many sequence learning applications including speech recognition and scene text recognition. However, CTC tends to produce highly peaky and overconfident distributions, which is a symptom of overfitting. To remedy this, we propose a regularization method based on maximum conditional entropy which penalizes peaky distributions and encourages exploration. We also introduce an entropy-based pruning method to dramatically reduce the number of CTC feasible paths by ruling out unreasonable alignments. Experiments on scene text recognition show that our proposed methods consistently improve over the CTC baseline without the need to adjust training settings. Code has been made publicly available at: https://github.com/liuhu-bigeye/enctc.crnn.

Keywords: Recurrent Networks · Regularization
[ Paper]  [ Poster] 

Author Information

Hu Liu (Tsinghua University)
Sheng Jin (Tsinghua University)
Changshui Zhang (Tsinghua University)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors