Machine learning turns data into models. This course covers supervised learning — regression, classification, decision trees, SVMs — unsupervised learning, neural networks, regularisation, and model evaluation.
Outcomes
- Train and evaluate supervised models on tabular data
- Detect and address overfitting using regularisation and cross-validation
- Implement gradient descent and explain its convergence behaviour
- Choose appropriate evaluation metrics for classification and regression
Outline
Start →- 01 Supervised Learning Regression and classification, the bias-variance trade-off, and the training/validation/test split.
- 02 Unsupervised Learning K-means clustering, hierarchical clustering, and PCA for dimensionality reduction.
- 03 Neural Networks Perceptrons, multilayer networks, backpropagation, and activation functions.
- 04 Model Evaluation Accuracy, precision, recall, F1, ROC curves, and cross-validation for reliable evaluation.
- 05 Regularisation and Optimisation L1 and L2 regularisation, gradient descent variants, and learning rate schedules.