MecaNano Live tutorial series on Machine Learning for Micro- and Nano-mechanics
by ,
MecaNano is the COST Action CA21121 "European Network for the Mechanics of Matter at the Nano-Scale", focused on nanoscale mechanics, nanomechanical testing, data interoperability, training, and the integration of experimental, modelling, and machine-learning approaches.
This online tutorial day introduces participants to the use of machine learning methods for the analysis, interpretation, and exploitation of nanoindentation and nanomechanical datasets. The event is designed as an interactive training activity, combining conceptual explanations with examples relevant to nanoindentation curves, high-throughput indentation maps, data-driven materials characterization, and physically meaningful model interpretation. The day includes two complementary tutorials.
Tutorials
Tutorial 1. Machine learning bases and advanced applications for nanoindentation data analysis
Edoardo Rossi
This tutorial introduces the foundations of machine learning for nanoindentation and nanomechanical data analysis. The session covers the basic concepts of supervised and unsupervised learning, feature extraction from indentation curves, clustering and classification of indentation datasets, analysis of high-throughput nanoindentation maps, and advanced workflows based on the full load-displacement curve. It discusses how data-driven methods can support phase identification, detection of anomalous curves, interpretation of mechanical populations, and integration with correlative microstructural information.
Tutorial 2. Explainable Machine Learning
Claus Trost
This tutorial focuses on explainable machine learning for materials mechanics and nanomechanical testing. The session discusses why explainability is essential when machine-learning models are used to analyse experimental datasets, where the results must remain physically meaningful and scientifically defensible. It introduces strategies to understand model decisions, identify relevant features, evaluate model reliability, and avoid black-box conclusions that cannot be connected to the underlying material behaviour or experimental conditions.
Programme
All times are given in the Europe/Zurich timezone. Each tutorial lasts two hours: 90 minutes of tutorial followed by 30 minutes of questions and discussion.
| 09:45 - 10:00 | Welcome and introduction to the MecaNano tutorial day |
| 10:00 - 11:30 | Tutorial 1: Machine learning bases and advanced applications for nanoindentation data analysis (Edoardo Rossi) |
| 11:30 - 12:00 | Questions and discussion |
| 12:00 - 14:00 | Lunch break |
| 14:00 - 15:30 | Tutorial 2: Explainable Machine Learning (Claus Trost) |
| 15:30 - 16:00 | Questions and discussion |
| 16:00 - 16:10 | Closing remarks |
Target audience
The event is intended for PhD students, postdoctoral researchers, and researchers working in nanoindentation, small-scale mechanical testing, materials characterization, and data-driven materials science. No advanced background in machine learning is required, although basic familiarity with nanoindentation data and scientific data analysis will be useful.
Learning outcomes
By the end of the tutorial day, participants should be able to:
- Understand the basic logic of supervised and unsupervised machine learning methods.
- Recognize how machine learning can be applied to nanoindentation curves and high-throughput indentation maps.
- Identify suitable workflows for clustering, classification, anomaly detection, and full-curve analysis.
- Understand the importance of interpretability when applying machine learning to experimental nanomechanics.
- Critically evaluate whether machine-learning outputs are physically meaningful and scientifically defensible.
Practical information
The tutorials will be held online. Connection details will be provided to registered participants before the event. Participants are encouraged to attend both tutorials, as the sessions are complementary. Any required material or additional instructions will be communicated through the event page.
Prof. Edoardo Rossi, Dr. Claus Trost