AI/ML for Fusion Summer School 2024#
The Open and Fair Fusion for Machine Learning Applications Project
(last update: June 19/2024)
Synopsis: An intensive 2-week summer school focused on undergraduate students with backgrounds in physics, engineering, computer science, applied mathematics and data science will be offered at William & Mary. This summer course will include a close to equal distribution of traditional instruction and active projects. The traditional instruction will provide daily 80 min instruction in morning classes with a focus on computing, applied mathematics, machine learning and fusion energy. These classes will be based on existing classes offered in data science at W&M, such as databases, applied machine learning and deep learning, Bayesian reasoning in data science. These classes will be supplemented with a class focused on fusion energy for the applications the students will tackle during the hands-on component and for students’ summer research. A draft agenda to be posted soon.
This course is based on the following references [D'Agostini, 2003, MacKay et al., 2003, Martin, 2018, Murphy, 2012, Pointer, 2019, Sutton and Barto, 2018]
Data Science (Lectures and Notebooks)
- (6/3/2024) - Intro to Data Science (get started, variance/bias) (slides, D. Vasiliu)
- (6/3/20224) - Intro to Data Science, Python Programming, Bias-Variance Tradeoff and the Gradient Descent Algorithm (D. Vasiliu)
- (6/3/2024) Supervised Learning, Linear Regression (D. Vasiliu)
- (6/4/2024) Regularization, Model Selection, Cross-Validation (D. Vasiliu)
- (6/4/2024) Supervised Learning - Logistic Regression (D. Vasiliu)
- (6/5/2024) Nonparametric Methods, GAMs (D. Vasiliu)
- (6/5/2024) Coding Examples Locally Weighted Regression (D. Vasiliu)
- (6/6/2024) Anomaly Detection, clustering with HDBScan (D. Vasiliu)
- (6/6/2024) Dimensionality Reduction, Unsupervised Learning (D. Vasiliu)
- (6/7/2024) Supervised Learning, Decision Trees, Random_Forest and XGBoost (D. Vasiliu)
- (6/7/2024) Intro to Deep_Learning, Multilayer_Perceptron (D. Vasiliu)
- (6/10/2024) Convolutional Neural Networks (D. Vasiliu)
- (6/10/2024) Graph Neural Networks (D. Vasiliu)
- (6/10/2024) Classification with C-Mod fusion data (J. Giroux)
- (6/10/2024) Regression with C-Mod fusion data (J. Giroux)
- (6/11/2024) Intro to Normalizing Flows (J. Giroux, slides)
- (6/11/2024) Normalizing Flows - Basic Example (J. Giroux)
- (6/11/2024) Normalizing Flows - GlueX BCAL Example (J. Giroux)
- (6/12/2024) A Gentle Introduction to Uncertainty Quantification and Bayes' Rule (slides, C. Fanelli)
- (6/12/2024) Sampling Techniques from Scratch (C. Fanelli)
- (6/12/2024) Introduction to Probabilistic Programming (coin example) (C. Fanelli)
- (6/12/2024) An Introduction to Bayesian Regression (slides, C. Fanelli)
- (6/12/2024) Bayesian Linear/Polynomial Regression - Basic Example (C. Fanelli)
- (6/12/2024) Bayesian Logistic Regression - Basic Example (C. Fanelli)
- (6/12/2024) Introduction to Bayesian Neural Networks (slides, C. Fanelli)
- (6/12/2024) Bayesian Neural Network - Basic Example (C. Fanelli)
- (6/12/2024) Exercise with CNN (D. Vasiliu)
- (6/12/2024) Exercise with GNN (D. Vasiliu)
- (6/13/2024) Intro to Optimization (slides, C. Fanelli)
- (6/13/2024) Single-Objective Bayesian Optimization (C. Fanelli)
- (6/13/2024) Multi-Objective Bayesian Optimization (C. Fanelli)
- (6/13/2024) Multi-Objective Genetic Algorithm (C. Fanelli)
Fusion Lectures
- (6/3/2024) W&M and Fusion (S. Mordijck)
- (6/3/2024) Nuclear Fusion Power - A solution to the world’s energy problem? (S. Mordijck, A. Dominguez)
- (6/4/2024) Overview of plasma diagnostics and measurements (E. Kostadinova)
- (6/5/2024) Alcator C-Mod (A. Saperstein, J. Stillerman)
- (6/6/2024) HDF (A. Jelenak)
- (6/7/2024) Fusion Pilot (S. Diem)
- (6/10/2024) Managing Data - Why it matters, when it is important, and how to do it (N. Cummings)
- (6/11/2024) Making plasma science open (N. Murphy)
- (6/12/2024) DIII-D ML/AI perspective (B. Sammuli)
- (6/13/2024) Data-mining the tokamak density limit (A. Maris)
Datasets
- C-Mod Dataset - variables explanation
- C-Mod Dataset - Additional Documentation
- Kaggle - Nuclear Fusion Data (Classification)
- Zindi - Multi-Machine Disruption Prediction Challenge for Fusion Energy by ITU (Classification)
- IOP paper - The updated ITPA global H-mode confinement database; description and analysis (Regression)
Additional resources