CSE762: Fairness, Ethics, and Explainability in AI

Foundations of Ethical AI: historical context, ethical principles, philosophical frameworks, legal regulations, case studies; Fairness in AI: defining fairness, sources of bias, detection and mitigation techniques, fairness in real-world applications, causal fairness; Explainability and Interpretability: importance, interpretability vs. explainability, XAI methods, post-hoc techniques, model-agnostic vs. model-specific approaches, evaluation metrics; Privacy and Security: privacy-preserving AI, data security, data governance, ethical data collection, differential privacy; Responsible AI Development: AI lifecycle, risk assessment, human-centered AI, social impact, AI auditing, governance frameworks; Emerging Challenges: AI autonomy, AI and creativity, disinformation, global AI ethics, future research directions.

Course Objectives

The core objectives of this course are to:
Recognize different types of biases in AI systems and analyze their societal impact.
Learn frameworks for ethical decision-making in AI development, including fairness, accountability, and transparency.
Explore methods for detecting, reducing, and preventing bias in machine learning models.
Develop techniques to improve model interpretability and explain AI decisions to stakeholders.

List of Books

1. To Be Added

Course Outcome