Course positioning
CS 281 is not pure philosophical discussion; it uses ML methods to address fairness, bias, transparency, and privacy. The course identifies unfair or untruthful problems through cases, applies detection and mitigation techniques, and discusses the formal limits and real trade-offs of these techniques.
What it covers
- Fairness: group/individual fairness, definitions and conflicts
- Bias detection: data bias, model bias, auditing
- Mitigation: pre-processing, regularization, post-processing
- Interpretability: attribution, counterfactuals, intrinsic interpretability
- Privacy: differential privacy, membership inference
- Formal limits: impossibility results, trade-offs
- Cases and project
Prerequisites
CS 229 or equivalent ML background.
Assessment & workload
Assignments plus a project/paper. Check the current syllabus for details.
Who it's for
People who want responsible AI, AI auditing, or policy technology; anyone who wants a hands-on ethics course.
Study advice
Pair with CS 139 for the human perspective and CS 221M for model internals.
Related courses
Official course description
Machine learning has become an indispensable tool for creating intelligent applications, accelerating scientific discoveries, and making better data-driven decisions. Yet, the automation and scaling of such tasks can have troubling negative societal impacts. Through practical case studies, you will identify issues of fairness, justice and truth in AI applications. You will then apply recent techniques to detect and mitigate such algorithmic biases, along with methods to provide more transparency and explainability to state-of-the-art ML models. Finally, you will derive fundamental formal results on the limits of such techniques, along with tradeoffs that must be made for their practical application. CS229 or equivalent classes or experience.