Title of the course: A crash course on Deep Learning; Theory and Practice
Instructor’s Name: Mr. Turan Bulmus
Institution: Google
Dates: 3-7 August 2020
Prerequisites: Basic Calculus and Linear Algebra, familiarity with optimization theory (unconstrained optimization), familiarity with python or R or any other programming languages.
Level: Graduate and advanced undegraduate
Abstract: Artifical Intelligence (AI) and Machine Learning (ML) are two concepts which attracted a lot of attention lately both in academia and also in industry. There is a lot of hype going on in different circles but a lot of these discussions are excluding the mathematical foundations behind ML and AI. With this course, we would like develop our AI/ML skills both for academic research purposes as well as commercial purposes. During this week long course we will focus on a popular set of algorithms of AI called deep learning. Deep learning is an neural network architecture which revolutionized the way we do machine learning with unstructured data (such as image, text and audio). We will start first mathematical foundations of and then further develop more complicated architectures focusing on different use cases. Depending on time constratints, we will go into Convolutional Neural Networks and Recurrent Neural networks. The topics will include how to build them, examples, optimizing architectures and open research questions. The lesson will also include hands on approach as well as examples of how these architectures are being used at commercial organizations.
Language: TR, EN