← Search

Pei Tang

2 accepted papers

2022

Domain Adaptation via Mutual Information Maximization for Handwriting Recognition

ICASSP 2022accepted

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 distributio…

Cited by 0SourceScholar
2019

Convolution with even-sized kernels and symmetric padding

NeurIPS 2019poster

Compact convolutional neural networks gain efficiency mainly through depthwise convolutions, expanded channels and complex topologies, which contrarily aggravate the training process. Besides, 3x3 kernels dominate the spatial representation in these models, whereas even-sized kernels (2x2, 4x4) are…