As biology is becoming more and more quantitative, today’s scientists end up with a huge amount of numbers to describe their experiments / their empirical observations. Traditional approaches, based on p-values and hypothesis testing, are very often pushed beyond their capabilities in these cases. In this 3 days workshop, we will cover the basics of machine learning (ML), namely how to extract information from datasets that could not be analyzed with the naked eye or manually. The aim is to share both the underlying mathematics (in a gentle way !) as well as provide a practical use of the methods, through dedicated softwares.
Students are more than welcome to come with their own datasets and/or share the ML methods they could have been already using. In that sense, the proposed schedule is only an outline and many of its parts could be covered by one or more willing participant. In the same spirit, if a specific method is of interest for a good number of people, it can be added in the program.