📚 Syllabus & Course Curriculum
Data Science & Analytics
Module-by-module breakdown of AI for Ocean Currents & Tsunami Early-Warning, from foundations to a certified capstone project.
Outline
Explore various ocean observations: Buoys/DART, HF radar, and satellite altimetry (SSH). • Understand data formats (NetCDF/Zarr), coordinates, and gridding for coastal domains. • Grasp physical basics of shallow-water intuition, bathymetry, and boundaries. • Compare ML landscape: baselines (persistence/AR) vs. advanced models like PINNs and Neural Operators (FNO).
Outline
Examine the essentials of Data Assimilation: EnKF/3D-Var intuition, analysis increments, and observation error. • Implement sequence models for marine time series data, including encoder–decoder transformers and SSMs, addressing masking and missing data. • Learn uncertainty and calibration techniques: ensembles, heteroscedastic outputs, and conformal prediction; assess coverage and CRPS. • Assimilate an HF-radar snapshot into yesterday’s state and visualize the resulting analysis increments.
Outline
Detect tsunami signals from DART/Tide Gauges: detrending, anomaly scoring, and ensuring robustness to noise/clock drift. • Estimate Tsunami ETA and communicate warnings effectively, including uncertainty bands and practical reporting. • Design alert mechanisms considering precision–recall vs. ROC, class imbalance, cost–loss analysis, and tiering (Advisory/Watch/Warning). • Detect tsunami-like anomalies and accurately estimate ETA with uncertainty bands.
Earn government-registered certification in AI for Ocean Currents & Tsunami Early-Warning
e-Certificate and e-Marksheet issued on successful completion.
