Title: Introduction to machine learning and its applications in biophysics and computational biology

Machine learning (ML) is a subset of artificial intelligence. Although the idea of “machine learning” can be traced back to the 1950s, it had not obtained the potential of becoming an extremely powerful and versatile method until the 1990s, when immense databases, so-called “big data”, started being available and thus the paradigm of ML research shifted from knowledge-driven approaches to data-driven approaches. Benefiting from the improvement in data collection and storage technology, large volumes of data have become accessible in more and more fields, and ML has been widely applied in various industrial sectors. In academia, ML’s potential for solving problems in different scientific disciplines has also attracted significant interest.

In this introductory presentation, the idea of “learning” in the context ML will first be introduced to give audience a brief idea about what ML is particularly good at. Technical challenges such as generalization, bias-variance tradeoff, and overfitting will be addressed, and the methods for overcoming such challenges, for example validation and regularization, will be discussed. Some most popular ML algorithms such as linear model, neural network, and decision tree will be introduced. Examples of ML research in computational biophysics and bioinformatics will be given.