Mathematics

Machine learning (ML) is receiving a huge amount of attention as a potential driver of discovery and innovation in an increasingly data rich world; however, the broad perception of ML, even within the scientific community, is often more the result of hype than reality. The purpose of this talk is twofold: in the first part, I will try to give a gentle introduction to ML, discussing its promise and limitations. Then I will move on to discuss how ML could be useful within a scientific context, and particularly in the life sciences, emphasizing the special constraints that using ML in a scientific biomedical environment is likely to require. Finally, I will give a recent example from my own work, on how we developed novel ML models to study RNA splicing in single cells.