Speaker: Dr. Narin Petrot, Naresuan University, Thailand
Email: [email protected]
Title: The path of the optimization algorithm development: from gradient descent to adaptive moment estimation methods
Abstract: In this talk, we will focus on the following main two parts: the algorithm and the model. In the first part, we discuss the motivations, pros, and cons of each famous optimization algorithm, which has been used in machine learning in the literature. While in the second part we will discuss the idea of developing an optimization model to avoid the important issues in machine learning such as overfitting phenomena.