Deep Learning for NLP brings forth new and unprecedented challenges. While deep learning has been employed very successfully in image processing, the NLP community is still coming to terms with the applications of deep learning methods in NLP. In this talk we would go over some applications of CNNs for NLP tasks. We would also explore other deep learning approaches which have worked well in NLP.

Convolutional Neural Networks have been applied to a number of NLP tasks with varying amount of success. The structural differences of text input vs that of images requires us to significantly change CNNs to suit NLP tasks. Some of these tasks such as Question Answering and Sentiment analysis can benefit greatly from CNNs. Apart from CNNs we will also look at models such as LSTMs and RNNs which are sequential models better suited to a large number of NLP tasks such as Machine Translation, Language Modelling etc.