Machine Learning Foundations
European certification validating fundamental skills in machine learning: supervised and unsupervised algorithms, data preprocessing, model evaluation and deployment.
Key skills
- ✓Implement supervised and unsupervised ML algorithms
- ✓Prepare and transform data for training
- ✓Evaluate and optimise model performance
- ✓Deploy ML models in a production environment
The question bank for this exam is prepared after your purchase. The exam is available no later than 48 hours after payment; you will receive an email as soon as it opens.
Is this right for you?
✓ This certification is right for you if:
- →Data analysts moving into machine learning
- →Python developers integrating predictive models
- →Data engineers new to ML
✗ This certification is not suitable if:
- →Beginners without Python or statistics basics: first consider 'Python for Data Analysis' Standard level
- →ML experts seeking advanced deep learning skills: consult 'Deep Learning Engineer' Master 801 level
Prerequisites & eligibility
Online registration, voucher payment, access to a computer with a webcam for monitoring.
Purchase options
Exam voucher, LMS pathway, bundle packs — build your order
Options d'achat
Composez votre parcours — voucher d'examen, préparation LMS, packs entreprise.
Skills assessed
ML Algorithms
Mastery of regression, classification, clustering and dimensionality reduction algorithms with scikit-learn and TensorFlow.
Feature Engineering
Advanced techniques for preprocessing, normalisation, encoding and feature selection to maximise model performance.
Evaluation & Tuning
Cross-validation, performance metrics, overfitting detection and hyperparameter optimisation with GridSearch and Optuna.
MLOps Foundations
Model versioning, CI/CD pipeline, drift monitoring and deployment with MLflow, Docker and REST API.
Exam format
Curriculum
Machine Learning Fundamentals
Introduction to learning paradigms (supervised, unsupervised, semi-supervised, reinforcement). Understanding the bias-variance tradeoff and its impact on generalisation. Train/validation/test split methodologies. Cross-validation (k-fold, stratified, time series split). Introduction to evaluation metrics based on problem type.
Preprocessing and Feature Engineering
Data exploration and analysis (EDA) with pandas and matplotlib/seaborn. Detection and handling of missing values (imputation, deletion). Management of outliers and aberrant values. Normalisation techniques (Min-Max, Z-score) and standardisation. Encoding of categorical variables (one-hot, label, ordinal, target encoding). Feature scaling and transformation (log, Box-Cox). Creation of derived features and interaction terms. Feature selection (filter, wrapper, embedded methods).
Supervised Algorithms
Simple and multiple linear regression, assumptions and diagnostics. Binary and multinomial logistic regression. Decision trees (CART) and split criteria (Gini, entropy). Ensemble methods: Bagging and Random Forest. Gradient Boosting (GBM, XGBoost, LightGBM, CatBoost). Support Vector Machines (SVM) for classification and regression. K-Nearest Neighbors (KNN) and choice of k. Introduction to neural networks: perceptron, MLP, activation function, backpropagation.
Unsupervised Algorithms and Evaluation
Clustering: K-means, choosing the number of clusters (elbow method, silhouette). DBSCAN and density-based clustering. Dimensionality reduction: PCA (Principal Component Analysis), t-SNE, UMAP. Classification metrics: accuracy, precision, recall, F1-score, support. Confusion matrix and interpretation. ROC curves and AUC. Regression metrics: MSE, RMSE, MAE, R². Detection of overfitting and underfitting.
Optimisation and Deployment
How to fund your certification
3 ways to cover the cost of your certification
OPCO — Employer funding
Ask your employer to cover the cost through your OPCO (employee training plan).
Personal — €450 incl. VAT
Secure online payment by credit card (Stripe).
Business rate: €585 excl. VAT
Business — custom quote
Teams, cohorts or private catalogue: a quote tailored to your volumes and goals.
Frequently asked questions
What is the difference with AWS ML or Google Cloud ML certifications?
PROVA Machine Learning Foundations is vendor-agnostic and focuses on the theoretical and practical fundamentals of ML, independent of cloud platforms.
Do I need to know deep learning for this certification?
No, this certification covers the fundamentals of classical machine learning. Neural networks are covered only in their basic principles (perceptron, simple MLP). For advanced deep learning (CNN, RNN, Transformers), consult our 'Deep Learning Engineer' Master 801 level certification.
Which languages and frameworks are used in the exam?
The exam is primarily based on Python with scikit-learn, pandas and numpy. Official documentation is accessible during the exam for specific syntax.
Is the certificate recognised by European companies?
Yes, PROVA is aligned with the EQF framework and referenced in ESCO, which guarantees its recognition in the European space.
Can I retake the exam if I fail?
Yes, you can retake the exam after a 14-day waiting period. A new voucher must be purchased. The PROVA DNA report from your first attempt precisely identifies your areas for improvement to optimise your preparation.
Your certification pathway
Before, during, after — the recommended progression
Droits du candidat
Transparence
Les critères de décision et le barème sont documentés et accessibles avant l'examen.
Droit d'appel
Toute décision peut être contestée dans les 30 jours. Examen par un tiers indépendant.
Plainte
Toute personne peut signaler un dysfonctionnement. Formulaire public accessible sans compte.
