Modelling The Language Of Life: From NLP to BioInformatics
Machine Learning Day
AI/Machine Learning/Deep Learning
Big Data Analytics
TimeWednesday, June 19th11:45am - 12:07pm CEST
DescriptionNatural language processing field has seen a remarkable breakthrough in 2018 in almost all public datasets by using the new era of context aware embedding, including different models like: ULMFiT, ELMo, OpenAI GPT and BERT. The advantages of these language models are that they can effectively capture the semantic, syntactic and grammar meaning of characters, words sentences from large unlabelled datasets. Which, can be used later to provide better representation for sentences for many NLP tasks. Unfortunately, these models have not been leveraged in other similar fields like Bioinformatics. This talk will give an overview of the current transfer learning context aware embedding models in NLP, and how it was applied in the Bioinformatics field to boost the performance on many use-cases. Furthermore, it will cover the current HPC limitation that makes it difficult to apply these techniques in the Bioinformatics field.