Learn about AutoML, the opportunities and challenges that arise in attempting to automate machine learning, and how this automation affects your organization.
Overview
Syllabus
Introduction
- How AutoML is changing analytics teams
- What you should know?
- What is AutoML?
- Understanding supervised machine learning on structured data
- Data engineering and ML Ops
- Understanding the ML lifecycle
- The challenge of ML problem definition
- Which phases have been automated most successfully?
- The challenge of automating data understanding
- What AutoML can and can't do during data prep
- AutoML's capabilities during the modeling phase
- Comparing model accuracy and business evaluation
- Monitoring and maintaining models
- The AutoML vendor landscape
- Demonstrating AutoML with KNIME
- A metaphor for AutoML
- Advice for team composition
- Next steps
Taught by
Keith McCormick
