Home / MLT
The algorithms: from PCA and clustering to SVMs, boosting and neural nets.
| Week | Lesson | Shows up in | Real practice |
|---|---|---|---|
| 1 | PCA: finding the best view of your data Introduction; Unsupervised Learning - Representation learning - PCA | Quiz 1 | 32 real questions |
| 2 | Kernel PCA: curved views of data Unsupervised Learning - Representation learning - Kernel PCA | Quiz 1 | 23 real questions |
| 3 | K-means: grouping without labels Unsupervised Learning - Clustering - K-means/Kernel K-means | Quiz 1 | 34 real questions |
| 4 | MLE, Bayes, Gaussian mixtures and EM Unsupervised Learning - Estimation - Recap of MLE + Bayesian estimation, Gaussian Mixture Model - EM algorithm | Quiz 1 | 33 real questions |
| 5 | Regression: least squares and its Bayesian view Supervised Learning - Regression - Least Squares; Bayesian view | Quiz 2 | 50 real questions |
| 6 | Ridge and LASSO: taming overfitting Supervised Learning - Regression - Ridge/LASSO | Quiz 2 | 18 real questions |
| 7 | k-NN and decision trees Supervised Learning - Classification - K-NN, Decision tree | Quiz 2 | 59 real questions |
| 8 | Generative models: Naive Bayes Supervised Learning - Classification - Generative Models - Naive Bayes | Quiz 2 | 12 real questions |
| 9 | Perceptron and logistic regression Discriminative Models - Perceptron; Logistic Regression | End Term | 39 real questions |
| 10 | Support vector machines Support Vector Machines | End Term | 33 real questions |
| 11 | Bagging, boosting and AdaBoost Ensemble methods - Bagging and Boosting (Adaboost) | End Term | 16 real questions |
| 12 | Neural networks Artificial Neural networks: Multiclass classification | End Term | 18 real questions |
Exam pattern and behaviour for MLT MLT formula sheet Past papers