Deep Learning

Course Number: 46937

This course introduces students to the foundations underlying deep learning that are relevant to the study of quantitative finance. While deep learning is a rapidly evolving field, this course focuses on the fundamental concepts that are pertinent to contemporary architectures and techniques. Topics include the basics of building deep neural networks; flexibility in modeling alternative types of data with convolutional neural networks; building recurrent neural networks (and related architectures) for autoregressive settings; as well as generative adversarial networks and autoencoders. There will be emphasis on the practical usage of techniques in quantitative finance with guided demos for implementation. Prerequisites: 46921, 46923, 46926, and 46927. Non-MSCF students may not take this course without written permission from the instructor. To be eligible, you must be a BSCF student, or a graduate student enrolled in an MSCF participating college/department (Stats & Data Science, Heinz, Tepper, Computer Science Dept.,or Math Sciences). PhD students with relevant research may be eligible with permission from the instructor, and you must satisfy the prerequisites.

Concentration: Statistics / Data Science
Semester(s): Mini 5
Required/Elective: Elective
Units: 6
Prerequisite(s): 46921, 46923, 46926, and 46927