Scaling Recurrent Neural Network Language Models
It addresses scaling challenges in language modeling for applications like speech recognition and machine translation, but is incremental in improving existing RNN methods.
The paper investigates the scaling properties of Recurrent Neural Network Language Models (RNNLMs), showing they achieve lower perplexities than n-gram models and deliver gains such as an 18% relative word error rate reduction on ASR tasks.
This paper investigates the scaling properties of Recurrent Neural Network Language Models (RNNLMs). We discuss how to train very large RNNs on GPUs and address the questions of how RNNLMs scale with respect to model size, training-set size, computational costs and memory. Our analysis shows that despite being more costly to train, RNNLMs obtain much lower perplexities on standard benchmarks than n-gram models. We train the largest known RNNs and present relative word error rates gains of 18% on an ASR task. We also present the new lowest perplexities on the recently released billion word language modelling benchmark, 1 BLEU point gain on machine translation and a 17% relative hit rate gain in word prediction.