The seminar presents an overview of the state-of-the-art approaches for optimizing compilers when machine learning techniques are applied. These techniques primarily enhance the quality of the obtained results and, more importantly, make it feasible to tackle two main compiler optimization problems: optimization selection (choosing which optimizations to apply) and phase-ordering (choosing the order of applying optimizations). The talk highlights the approaches taken so far, the obtained results, the fine-grain classification among different approaches and finally, the most influential papers in the field.

Dr. Amir H. Ashouri is a postdoctoral researcher at the Department of Electrical and Computer Engineering at University of Toronto, Canada. Prior to joining the University of Toronto, he completed his M. Sc. (2012) and Ph.D. (2016) at Politecnico di Milano under the supervision of Prof. Cristina Silvano and Prof. Gianluca Palermo. He has been researching on deep learning, compiler optimizations, and automatic tuning techniques using machine learning.His Ph.D. thesis was selected as the winner of IEEE-Italy section best Ph.D. thesis of 2016. Recently, a book entitled "Automatic Tuning of Compilers Using Machine Learning" has been adapted from his Ph.D.