5-day workshopAI & DataIntermediate

Machine Learning for Materials Science

Apply machine-learning methods to materials data — from featurisation and property prediction to model interpretation and materials discovery workflows.

Syllabus

  1. Introduction to data-driven materials science
  2. Materials databases, data cleaning and featurisation
  3. Supervised learning: regression and classification for property prediction
  4. Model validation, uncertainty and interpretability
  5. Hands-on mini-project and presentations

Learning outcomes

By the end of this programme, participants will be able to:

  • Build and curate materials datasets from open databases
  • Generate composition- and structure-based descriptors
  • Train, validate and interpret regression and classification models for material properties
  • Understand the role of ML in high-throughput screening and materials discovery
  • Complete a mini-project on a real materials dataset

Prerequisites

Basic Python programming and undergraduate-level chemistry, physics or materials science.

Instructor

Instructor details will be published with the programme schedule. See our team page for faculty profiles.

Certificate policy

Participants who attend at least 80% of sessions and complete the required assignments or project receive a Certificate of Completion from Crystal Bridge – Education & Research. Certificates are issued digitally after the programme ends.