Since the beginning of COVID outbreak, numerous papers have been written describing related medical research, many of which are available in CORD dataset. In this talk, we will describe our experiment of applying Open Domain Question Answering model to this dataset using recent BERT model available in DeepPavlov NLP framework. We will also demonstrate how this model can be trained in a cost-effective manner on Azure Machine Learning, using different compute options for data preparation and training.
Mikhail Burtsev Ph.D.
Head of Neural Networks and Deep Learning Lab at MIPT, Founder of DeepPavlov
Experienced Head of Deep Learning research projects with a demonstrated history of working in the research institutions. Skilled in Neural Networks, Mathematical Modeling, Python, Computer Science, Cognitive Science, and Genetic Algorithms. Strong research professional with a Doctor of Philosophy (Ph.D.) focused in Artificial Intelligence.
Dmitry Soshnikov Ph.D.
Cloud Developer Advocate at Microsoft, Associate Professor at MIPT
Dmitry is Cloud Developer Advocate for Microsoft, specializing in AI/ML and F#. He also teaches AI and Functional/Logic Programming at top Moscow Universities. Passionate father, Digital Artist, Gun Fu Cha master, etc.
Local ODSC chapter in Dublin, Ireland
Use discount code - Meetup2020 - to get extra 10% off on your pass for Virtual Conference West and Virtual Conference APAC.
BERT-based Open Domain Question Answering using DeepPavlov Library on Azure Machine Learning
AI+ Subscription Plans
UPCOMING LIVE TRAINING
Register now to save 30%
All Live Training
PAST LIVE TRAINING: Available On-Demand:: Modeling and Machine Learning in Theory and Practice3 Lessons $244.86
All Live Training
PAST LIVE TRAINING: Available On-Demand:Animating Data: Turn any Matplotlib plot into an animated gif(1) 5.0 average rating1 Lessons $147.00
All Courses, All Live Training
PAST LIVE TRAINING: Available On-Demand: Gradient Boosting for Prediction and Inference(3) 5.0 average rating2 Lessons $178.08