VTR UGE 21 value added course, Vel Tech University, Chennai
Prof. Dr. Utku Kose
A five-day course on explaining machine learning models, with lecture pages, animations, interactive labs, Colab notebooks and a Python track for beginners. Each day starts from its overview, which gives the study path and links to the lecture page, the Colab notebook, the interactive lab and the PDF notes. Progress is stored only in this browser.
Why models need explanations, the map of explanation methods, models that explain themselves and a model that is accurate for the wrong reason.
Day overview | Lecture | Colab notebook | Interactive lab | PDF
Explaining any model from its predictions alone: which features matter, how they act, why one case got its prediction and what would change it.
Day overview | Lecture | Colab notebook | Interactive lab | PDF
Where a network looks: gradients, SmoothGrad, Integrated Gradients and Grad-CAM (gradient-weighted class activation mapping), checked against a known truth and a sanity test, and geodesic explanations with GEMEX (Geodesic Entropic Manifold Explainability).
Day overview | Lecture | Colab notebook | Interactive lab | PDF
When can a probability, an explanation and a decision be trusted? Calibration, shift, manipulated explanations, fairness and a benchmark against random.
Day overview | Lecture | Colab notebook | Interactive lab | PDF
From explanations to artefacts that people use: records, dashboards, model cards, risk tiers, and three case studies from finance, maintenance and text.
Day overview | Lecture | Colab notebook | Interactive lab | PDF