AI for Portfolio Optimization: Beyond Classical Mean-Variance

Classical portfolio optimization gave investors a powerful framework, but it also exposed the fragility of any approach that depends on uncertain inputs. AI does not eliminate that uncertainty, but it changes how investors estimate it, model it, and react to it. This course gives you a calm, structured introduction so you can speak confidently about AI in portfolio work without overstating its powers. You will learn how classical optimization works, where it tends to break, and how modern methods including risk parity, hierarchical risk parity, and machine learning extensions address those weaknesses. The course stays grounded in widely used concepts and respects the realities of investment management. What you'll learn: - Understand classical mean-variance optimization and its sensitivity to input estimates - Recognize the appeal of risk parity and hierarchical risk parity as alternative weighting frameworks - Explore how machine learning estimates returns, covariances, and regime changes more robustly - Read how AI supports tactical allocation, factor exposure analysis, and rebalancing decisions - Identify the risks of AI in portfolio work including overfitting, regime change, and false confidence - Understand the integration points between AI-supported optimization, execution systems, and risk management The course begins with classical optimization, moves through modern alternatives and AI extensions, and closes with the operational realities of running AI-supported portfolios. Written exercises help you connect each concept to a realistic portfolio or asset class. This course is designed for absolute beginners with no portfolio theory or AI background, including finance students, investment operations professionals, and software developers entering quantitative finance. No prerequisites are needed beyond general comfort with mathematics. The course is informational and does not provide investment advice; it builds the literacy that lets you ask better questions.