Laboratory of Neural Systems and Deep Learning
Authorization required.
The Laboratory of Neural Systems and Deep Learning at MIPT conducts research in the field of deep neural network architectures for working with natural language text (NLP & Conversational AI), as well as in the fields of deep learning theory and neural architecture search.
- Reinforcement learning
- Natural Language Processing (NLP)
- Machine learning
Mikhail Burtsev
Head of Laboratory
Research directions
Knowledge Graph-Based Dialogue Generation
+
Responses can be more meaningful compared to chit-chat generated responses thanks to usage of facts coming from a knowledge base like Wikidata. The system consists of three components: extraction of triplets from Wikidata for entities from the user's utterance; triplets ranking (choosing the triplet which is the most appropriate to use for generation of the response utterance); generation of the response utterance.
New Transformer Architectures
+
We are working on memory augmented transformer-based language models, knowledge graph integration to language models, knowledge distillation and we are experimenting with transformer architecture modifications: multiple streams, bottlenecks, sentence-level representations.
Publications and patents
Lab address
Долгопрудный, Институтский переулок, 9
Authorization required.