BIOE 488

Credit: 3 hours.

Will introduce students to basic principles and practical applications of scientific computing as they relate to problems in machine learning and computed imaging. In this self-contained course, students will be introduced to a variety of important topics that underlie real-world machine learning and biomedical image computing tasks that are not typically comprised in a single course. The material will be presented in a practical way that will be accessible to engineering students who have a moderate level of experience in scientific computing but lack specific training in computer science. The emphasis will be on immediate applicability of scientific computing techniques as opposed to theoretical knowledge. The course will begin with an overview of good scientific coding practices in Python and introductions to canonical computing architectures. Subsequently, parallel computing concepts will be surveyed that address multi-core CPU and GPU-enabled systems. Distributed GPU computing on a cluster will also be covered. The salient aspects of TensorFlow and/or other relevant machine learning programming environments will be introduced and utilized in applications of machine learning.

3 undergraduate hours. 3 graduate hours. Prerequisite: Familiarity with the Python programming language. Restricted to students with senior undergraduate or graduate standing in an engineering major.