AI Safety, Alignment and Evaluation
Making AI systems do what we intend: specification, robustness, interpretability and evaluation.
AI-430 Year 4 3 credits 5 lessons planned
Offered by HSM University · Artificial Intelligence
Course content
- 1Why alignment is hardComing
- 2Evaluating capabilities and risksComing
- 3Robustness and adversarial attacksComing
- 4InterpretabilityComing
- 5Oversight of advanced systemsComing
Requirements
Recommended background. You can start any course at any time.
Description
Making AI systems do what we intend: specification, robustness, interpretability and evaluation.
Part of the Science stream, typically taken in year 4.