Week 1
- Responsible AI
- Correctional Offender Management Profiling for Alternative Sanctions
- CORELS
- Fairness Impossibility Theorem
- Fairness
Week 2
- Black Box
- Explainable Machine Learning
- Epistemic Opacity
- Believing in Black Boxes
- Rashomon Effect
- Differential Privacy
- Differential Privacy Stochastic Gradient Descent
- Model Privacy Performance Trade-off
- GDPR Right to be Forgotten
- Machine Unlearning
Week 3
- AI Governance
- Forms of Governance
- Directive on Automated Decision Making
- Quebec Law 25
- Real Risk of Significant Harm
- Artificial Intelligence and Data Act
- Human Centered AI
- AI Alignment
- AGI
- TAI
- Misalignment
- Instrumental Convergence Hypothesis
- Orthogonality Thesis
- Arguments Against Alignment
- RLHF
- Supervised Fine Tuning
- Constitutional AI
- Scalable Oversight
- Adversarial Training
- Mechanistic Interpretability