Diploma in Technology Management and Entrepreneurship
| TME6015 | AI/ML Workflow Design | 3 ch |
|---|---|---|
The purpose of this course is to engage broad audiences in Engineering and Science for efficient design of AI/ML workflows in different application scenarios. Students will learn how different ML models can be designed and fitted into efficient data structures for representational learning. The course starts with data-centric approach all the way to the model-centric approach designs. In data-centric we will cover topics in supervised labeling, active learning, transferring expert domain knowledge into supervised labels and annotations, statistical analysis of supervised data and their class representation. In model-centric approach we will cover broad topics of supervised and unsupervised machine learning models in details. The general overview of deep learning will be introduced and how we can use different ML models as plug-play tool to fit the labeled data for training purposes. Techniques of optimization and hyper-parameter settings will be studied. Popular applications in deep learning will be introduced in the context of audio classification, image classification, tabulated data classification, and time-sequence data classification. Prerequisites:
| ||
