Is it possible to build machine-learning models without machine-learning expertise? BioAutoMATED, an open-source, automated machine-learning platform developed by Jim Collins, life sciences faculty lead and colleagues at the MIT Jameel Clinic, the epicentre of machine learning and health at the Massachusetts Institute of Technology (MIT) that whittles down a months-long process of data model selection and processing to just a few hours.
In any machine learning project, selecting the appropriate model, formatting the dataset for the model and fine-tuning it can dramatically change how the model performs, and requires a significant amount of time and work. This is a prohibitive step for a lot of biology researchers who want to use machine learning, which is a tool currently under-used in subfields of biology. BioAutoMATED aims to help in this regard, by democratising artificial intelligence for research labs.