Meta-Learning a Dynamical Language Model
arXiv:1803.10631
Abstract
We consider the task of word-level language modeling and study the possibility of combining hidden-states-based short-term representations with medium-term representations encoded in dynamical weights of a language model. Our work extends recent experiments on language models with dynamically evolving weights by casting the language modeling problem into an online learning-to-learn framework in which a meta-learner is trained by gradient-descent to continuously update a language model weights.
arXiv comment5 pages, 2 figures, accepted at ICLR 2018 workshop track
References in corpus (11)
- Neural Machine Translation by Jointly Learning to Align and Translate
- Overcoming catastrophic forgetting in neural networks
- Pointer Sentinel Mixture Models
- Regularizing and Optimizing LSTM Language Models
- Learning to Generate Reviews and Discovering Sentiment
- Learning to learn by gradient descent by gradient descent
- Hierarchical Multiscale Recurrent Neural Networks
- Using Fast Weights to Attend to the Recent Past
- Learning to Optimize
- Memory-Efficient Backpropagation Through Time
- Recurrent Memory Networks for Language Modeling