Byte-based Language Identification with Deep Convolutional Networks

Abstract

We report on our system for the shared task on discriminating between similar languages (DSL 2016). The system uses only byte representations in a deep residual network (ResNet). The system, named ResIdent, is trained only on the data released with the task (closed training). We obtain 84.88% accuracy on subtask A, 68.80% accuracy on subtask B1, and 69.80% accuracy on subtask B2. A large difference in accuracy on development data can be observed with relatively minor changes in our network's architecture and hyperparameters. We therefore expect fine-tuning of these parameters to yield higher accuracies.

7 pages. Adapted reviewer comments. arXiv admin note: text overlap with arXiv:1609.07053

References in corpus (3)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• Semantic Tagging with Deep Residual Networks

Cited by in corpus (1)

• Semantic Tagging with Deep Residual Networks