Course Description

Machine learning and AI present an opportunity to scale and generalize many minor decisions that will automate our lives and make us more effective at our work. From automatically determining if an employee has access to resources to what is shown on our news feed to self-driving vehicles, and beyond to public safety - AI is becoming a daily part of our lives. As a result, we have begun to see the problems that AI can introduce, from propagating gender or racial biases to creating deep fakes and furthering divisive arguments online. Machine learning and AI are therefore no longer simply a technical or scientific exercise, data science practitioners must consider the ethical implications of the models they train. 

It is tempting to think that the statistical nature of machine learning and AI ensures that they behave purely mathematically and without the emotion or biases of human decision-makers. Models, however, are trained and tuned by data scientists on data collected and created by people, therefore they often are mirrors that reflect our worst attributes. In this course, we confront the contradiction between training a model that reflects real-world bias and biasing it purposefully to have it behave differently, even if more fairly. Through case studies and discussions, this course is designed to allow students to consider fully the ethical questions that arise in an AI context and identify practices that ensure they deploy responsible data products.

This course is part of the Ethics and Leadership track of the Advanced Data Science Certificate.

Course Objectives

Upon successful completion of the course, students will:

  • Discuss and identify how bias affects machine learning models.

  • Explore the influence of machine learning on decision making.

  • Identify case studies where machine learning has been the subject of ethical gray areas.

  • Describe fairness in a mathematical or scientific context.

  • Identify practices that data scientists should follow to ensure the ethical construction of machine learning models.

  • Discuss practices and workflows that can eliminate biases from machine learning applications.

Applies Towards the Following Certificates

Thank you for your interest in this course. Unfortunately, the course you have selected is currently not open for enrollment. Please complete a Course Inquiry so that we may promptly notify you when enrollment opens.
Required fields are indicated by .