From the course: LinkedIn AI Academy AI-100: 2 Supervised Learning with Neural Networks

Unlock the full course today

Join today to access over 24,500 courses taught by industry experts.

Recurrent neural network language models

Recurrent neural network language models

- [Instructor] Let's talk about a new kind of neural network called the recurrent neural network which is used for doing language modeling and other sequence classification tasks. We build up the idea of recurrent neural networks using the same task we described in the last video that of predicting the next word in the sentence given all the previous words. As before let's consider the sentence, "How old are you?" Each word represents a different point in time. The word how occurs at time instant one. The word old occurs at time instant two, and so on. Given the first word of the sentence how, we want our network to predict the next word of the sentence, old. Since there's no previous context before the first word how, a feed forward neural network can accomplish this task. So we move on to the next word. Now the feed forward neural network has to predict the word are because the feed forward neural network can only…

Contents