COMP 685 - LEARNING & ALGORITHMS FOR COMP
Long Title: LEARNING AND ALGORITHMS FOR COMPUTATIONAL MEDICINE
Department: Computer Science
Grade Mode: Standard Letter
Language of Instruction: Taught in English
Course Type: Seminar
Credit Hours: 3
Description: Hospitals continuously collect vast and ever increasing volumes of medical data. This
data provided unique opportunities that could improve human health, detect diseases in
early stages, and develop personalized treatments. This creates a rapidly changing field
of computational medicine with a plethora of algorithms and machine learning methods
as well as new and exciting research challenges. In this class we aim to study some of
the most advanced methods for computational medicine and look at these challenges
from a machine learning and algorithms perspective.
We will focus on novel and unique challenges of computational medicine such as
privacy, fairness, efficiency, explainability, generalization, learning from multiple
modalities, federated learning and analytics, continual learning and so on. The course
will specifically focus on methods with provable guarantees such as differential privacy,
convergence analysis, algorithmic fairness and so on. Recommended Prerequisite(s): COMP 582