Generating Synthetic Data With GANs
Do we have enough data? Are our datasets imbalanced? How can we accelerate research while avoiding data leakage?
Please register using the zoom link to get a reminder:
Do we have enough data? Are our datasets imbalanced? How can we accelerate research while avoiding data leakage? Generative Adversarial Networks (GANs) are a promising AI solution to these questions. With GANs, we can generate more and better data that is more fair and generalizable, which can be used to improve ML models and algorithm testing. GANs are also promising in the field of data privacy, since they could break barriers to data sharing, allowing companies and institutions to accelerate research findings.
11:45 am – 11:55 am Arrival, socializing and Opening
11:55 am – 1:00 pm Marta Batlle López, “Generating Synthetic Data With GANs”
1:00 pm – 1:10 pm Q&A
About Marta Batlle López
As a data scientist in Pharma Informatics and Product Development at Roche, Marta is currently leading the development of a Deep Learning group focused on Generative Adversarial Networks and their applications in healthcare. In the past, she has also worked on different projects in the field of Natural Language Processing to speed up clinical trials. She is currently a One Young World Ambassador for Roche, with the goal to promote AI fairness within the company. Prior to joining Roche, she completed an MSc in Health Data Science and an MRes in Neuroscience at University College London.