Data Science and Machine Learning Design Laboratory

4 credits · College of Engineering catalogue

Advisory Prerequisite: ECE 505 or ECE 551 or graduate equivalent. Enforced Prerequisite: None. (4 credits) This course uses a sequence of hands-on projects to bring into sharper focus the following concepts in the data-to-decision cycle: 1. how smart (or bad) data can positively (or negatively) affect decisions in the design and operation of an engineering system; 2. how to acquire such data, clean and store it via appropriate pre-processing and post-processing it for aiding reproducibility; 3. how to display, render, deploy and interpret it in the context of a real or simulated close-up loop type cloud based engineering system; and finally, 4. how to communicate the shortcomings and vulnerabilities of such systems, including plug-and-play systems using pre-trained off-the-shelf deep learning models, when integrated into a decision-making system; 5. conceptualization and execution of an open-ended, reproducible cloud-based design project.