📚 Syllabus & Course Curriculum
Immunology & Vaccine Development
Module-by-module breakdown of ML for Epidemiological Forecasting and Vaccine Logistics Optimization, from foundations to a certified capstone project.
Surveillance
• Case, hospitalisation and death series and their differing reliability
• Reporting delays, backfill and weekday effects that mislead naive models
• Nowcasting the present before attempting to forecast the future
Mechanistic
• SIR and SEIR structure, R0 and the effective reproduction number
• Parameter estimation and identifiability problems with limited data
• What a mechanistic model gives that a pure ML model cannot: interpretable scenarios
Statistical
• Time series methods, gradient boosting and sequence models on epidemic data
• Feature construction from mobility, weather and search data, with their instability
• Hybrid and ensemble forecasts, which consistently beat single models
Evaluation
• Probabilistic forecasts, quantiles and the weighted interval score
• Backtesting with data as it appeared at the time, not as later revised
• Communicating uncertainty to decision makers without collapsing to a point
Logistics
• Cold chain constraints, wastage and the shelf life of a thawed vial
• Facility location, allocation and routing as optimisation problems
• Equity constraints in allocation and last-mile reality in rural distribution
Earn government-registered certification in ML for Epidemiological Forecasting and Vaccine Logistics Optimization
e-Certificate and e-Marksheet issued on successful completion.
