Scientific SecretaryConsultants' Meeting on Machine Learning for Nuclear Data
Purpose of the Meeting
The purpose of this meeting is to get an overview of work already undertaken at the intersection of machine learning and nuclear physics, to explore to which extent methods of machine learning and statistics (e.g., outlier detection, quality assurance, inference of typical shapes of cross section curves, uncertainty quantification) can help us to get a better understanding of nuclear data (e.g., in databases, by meta-analyses of various experiments) and identify potential directions of development.Agenda
The agenda is available here.
Summary Report
The summary report INDC(NDS)-0823 is available.
Presentations
| # | Author | Title | Link |
| 1 | G. Schnabel | Introduction to Consultancy Meeting on Machine Learning for Nuclear Data | pptx |
| 2 | A. Gray | Bayesian Calibration with Transitional MCMC | pdf | pptx |
| 3 | A. Gray | When marginals or dependencies are unknown: computing with imprecise probabilities | pptx |
| 4 | D. Neudecker | Validating Nuclear Data Augmented by Random Forests | |
| 5 | D. Neudecker | Highlighting Physics Reasons for Discrepancies in Differential Experimental Data via Elastic Net and Random Forest | |
| 6 | M. Grosskopf | Using Machine Learning to Explore, Diagnose, and Correct for Bias in Nuclear Data | |
| 7 | F. Caliva | Anomaly detection in nuclear reactors using deep learning | |
| 8 | F. Bachoc | Gaussian processes under inequality constraints | |
| 9 | T. Kin | Machine learning in radiation metrology: Neutron spectrum unfolding and Gamma-ray spectrometry | pptx |
| 10 | E. Alhassan | Towards the inclusion of model uncertainties in nuclear data evaluations | |
| 11 | H. Sjöstrand | Gaussian Processes for treatment of model defects in nuclear data evaluations | pptx |
| 12 | V. Sobes | AI/ML-based evaluation in the resonance region | |
| 13 | G. Pleiss | GPyTorch: Gaussian Processes for Modern Machine Learning Systems | |
| 14 | H. Iwamoto | Nuclear data generation using Gaussian process regression and its related topics | pptx |
| 15 | G. Schnabel | Uninformative, sparse, smooth: GP priors on second derivatives of cross sections | |
| 16 | N. Dwivedi | Trees, Forests and Islands: A Machine learning approach to Nuclear Physics | pptx |
| 17 | D. Brown | Machine Learning for Neutron Resonance Evaluations | |
| 18 | A. Lovell | Probabilistic Machine Learning for Uncertainty Quantification | |
| 19 | D. Siefman | Constrained Bayesian Optimization of Criticality Experiments at LLNL | pptx |
| 20 | G. Schnabel | Opening of final day of CM on ML for ND | pptx |