The goal of the project is to compare empirically the generalization error of two stochastic optimization algorithms used in Reinforcement Learning (SCRN and momentum-based SGD) on cost functions that satisfy the so-called gradient dominance property.

Requirements: Strong Python programming skills (preferred) Experience with ML libraries such as Numpy, Pytorch Knowledge about elementary machine learning theory If interested, please send your CV and a transcript of your grades to mohammadsaeed.masiha@epfl.ch