Deep learning in natural language processing

MIPT Laboratory of Neural Networks and Deep Learning opens the Deep Learning in Natural Language Processing course . It starts on February 21 and will be held every Wednesday at 19:00 in 105 BC, you need to register . The duration of the course is 10 lessons, they will be conducted by a laboratory employee Valentin Malykh ( madrugado ). The course is open to all who wish to deepen their knowledge in Deep Learning and Natural Language Processing. Classes will also be broadcast online in the lab’s Youtube channel .

The program is based on a similar course at Stanford University. Training will consist of lecture viewing, quiz answers, homework and project. The seminars will be devoted to the discussion of difficult moments in lectures, the analysis of quizzes and homework, as well as consultations on projects.

Classes will be held in the following order:

1. Intro to NLP and Deep Learning;
2. Simple Word Vector representations: word2vec, GloVe;
3. Advanced word vector representations: language models, softmax, single layer networks;
4. Word Window Classification and Neural Networks;
5. Backpropagation & Practical Tips;
6. Recurrent Neural Networks and Language Models;
7. Machine translation and advanced recurrent LSTMs and GRUs;
8. Neural Machine Translation and Models with Attention;
9. Convolutional Neural Networks;
10. Dynamic Neural Networks for Question Answering.

Online information about the course can be tracked in the telegram channel .