Brief Announcement: Gradual Learning of Deep Recurrent Neural Network Conference Paper uri icon

abstract

  • Abstract Deep Recurrent Neural Networks (RNNs) achieve state-of-the-art results in many sequence-to-sequence modeling tasks. However, deep RNNs are difficult to train and tend to suffer from overfitting. Motivated by the Data Processing Inequality (DPI) we formulate the multi-layered network as a Markov chain, introducing a training method that comprises training the network gradually and using layer-wise gradient clipping. In total, we have found that applying our methods combined with previously introduced regularization and optimization methods resulted in improvement to the state-of-the-art architectures operating in language modeling tasks.

publication date

  • January 1, 2018