Click here to book your place. Places are limited.
Book nowTuesday 14 April - Thursday 16 April 2026
In-person at Burlington House, Piccadilly London
The course will run from 9am until 5pm each day.
Join Professor Cédric M John from Queen Mary University of London (QMUL) for an intensive three-day in-person course featuring 21+ hours of hands-on training in data science and machine learning.
This course provides a rigorous and hands-on introduction to the foundational principles, tools, and workflows of modern machine learning. Aimed at participants with a background in science or engineering, it emphasises not only the mathematical underpinnings of key algorithms, but also the practical skills required to implement, evaluate, and improve predictive models using real-world datasets.
Throughout the course, participants will explore the entire machine learning lifecycle - from data preprocessing and exploratory analysis to algorithm selection, model tuning, and performance evaluation. The course takes a model-centric approach, gradually building up from linear models to advanced ensemble techniques and unsupervised learning. Participants will gain familiarity with essential Python libraries such as scikit-learn, and learn best practices for developing reproducible, modular code using workflows and pipelines.
In the latter part of the course, participants will explore deep learning and unsupervised learning techniques, equipping them to handle complex, high-dimensional, or unlabelled data. By the end, participants will be capable of designing, implementing, and critically assessing machine learning solutions to a wide range of applied scientific problems.
Please note: This course is not CPD accredited
- Preprocess and structure data for use in machine learning pipelines, including handling missing values, feature engineering, and normalisation.
- Conduct exploratory data analysis (EDA) to assess data quality, identify trends, and inform algorithm choice.
- Explain and implement core supervised learning algorithms, including linear regression, logistic regression, k-nearest neighbours, support vector machines, decision trees, and ensemble methods.
- Evaluate model performance using appropriate metrics for both regression and classification tasks (e.g., RMSE, accuracy, F1-score, ROC-AUC).
- Understand the mathematical foundations of model fitting, including gradient descent and common loss functions.
- Use model optimisation and tuning techniques, such as cross-validation, hyperparameter tuning, and grid/random search strategies.
- Build modular, reproducible ML workflows using Python classes and scikit-learn pipelines.
- Apply advanced modelling techniques such as ensemble methods (bagging, boosting, stacking) to improve predictive accuracy.
- Deploy unsupervised learning methods, including dimensionality reduction (PCA), clustering (KMeans, DBSCAN), and anomaly detection techniques.
- Critically assess the limitations of machine learning models, including issues of overfitting, bias-variance trade-off, and the implications of the No Free Lunch Theorem.
- Design and document a complete machine learning project, from data ingestion to model deployment, using clean and maintainable Python code.
This course is aimed at professional or academic geologists with some notion of coding – preferably in Python – who are looking for an intensive first course in data science and machine learning. This is also relevant for graduate students who want to learn the ropes and apply machine learning methods into their research project.
Click here to book your place. Places are limited.
Book nowTentative Programme:
The course will run from 9am until 5pm each day.
Tuesday 14 April: Python for data science
An introduction / review of coding in Python for data science; Although participants are require to have some good knowledge of Python, we will dedicate one day on this to instil a solid foundation for the rest of the course.
Wednesday 15 April: Machine Learning: Fundamentals
Fundamental concepts in data science and intro to Machine Learning; This will also include introducing a few ML algorithms on that day, as well as the general concept of how to use them.
Thursday 16 April: Deep Learning Neural Networks and more
Deep Learning: on the last day, we will focus on neural networks, and introduce a few complex architectures such as CNNs since images are key in geosciences The course will be a mix of lectures, and practical exercises
Click here to book your place. Places are limited.
Book now
Professor Cédric M. John
Professor Cédric M. John is the Head of Data Science for the Environment and Sustainability at the Digital Environments Research Institute (DERI) at Queen Mary University of London. Previously, he was appointed as Reader in Earth-Centric AI at the Department of Earth Science and Engineering, Imperial College London (2008-2023).
Professor John's research focuses on applying AI and machine learning to subsurface data, such as core images, logs, seismic, and geochemical datasets. His work includes automatic facies classification, generative AI for core image reconstruction, and deep learning for seismic data analysis. He has developed and taught undergraduate and master level courses like "Data Science and Machine Learning for Geoscientists" and "Data Science and Machine Learning for Planet Earth," and “Advanced Carbonate Reservoirs”. At Imperial, he was co-creator and deputy director for the “Geo-Energies with Machine Learning and Data Science” MSc program.
Throughout his career, Professor John has supervised numerous PhD students and postdoctoral researchers. Much of his research has been industry funded, and he has work closely on technologies to promote the energy transition. His long-term ambition is to advance the integration of AI in geosciences through his research and teaching.
Click here to book your place. Places are limited.
Book nowEarly-bird fees (until 14 February 2026)
Fellow £1,349
Non-Fellow £2,699
Student Member £1,079
Student Non-Member £2,159
Corporate Patron £1,349
GSA Member £1,619
Partner £2,294
Standard fees (after 14 February 2026)
Fellow £1,499
Non-Fellow £2,999
Student Member £1,199
Student Non-Member £2,399
Corporate Patron £1,499
GSA Member £1,799
Partner £2,549
The Geological Society of America (GSA) members discount
We offer a generous 40% discount off our Non-Fellow rate to members of the Geological Society of America (GSA). A discount code must be quoted on the registration form in order to take advantage of the discount. If you are a member of GSA and do not have the discount code, please email training@geolsoc.org.uk with proof of your membership.
Group Discount
5+ delegates: 10% off
For more information on how to access the Group Discount, please email training@geolsoc.org.uk
Concessions
We offer students a generous discount. Please verify your student status by either registering with your student email address, or uploading a photograph of your student identification/acceptance letter.
If you require an invoice to register for this course, please email training@geolsoc.org.uk
Registration will close 24 hours before the event takes place.