DOMAIN ADAPTATION VIA MUTUAL INFORMATION MAXIMIZATION FOR HANDWRITING RECOGNITION

Deep learning models for handwriting recognition have been developed in recent years. To improve the model's generalization ability for sequence modeling task, this paper proposes to use domain adaptation with statistical distribution alignment and entropy regularization. For statistical distribution alignment, a domain adaptation loss function is proposed by using both the first and second order statistical information of deep feature representations, which is equivalent to maximizing the mutual information in feature spaces of the source domain and target domain. For entropy regularization, the entropy of the predicted text symbols of unlabeled samples in the target domain is also utilized as an additional loss function, which actually maximizes the mutual information between the feature space and pattern space in the target domain. The experimental results on the IAM handwriting dataset have demonstrated the effectiveness of the proposed domain adaptation method for sequence modeling task.