Term: | Hilary Term 2017, Jan 15 - Mar 11 |

Lectures: | Statistics LG.01, Wed 12-13, Thu 16-17 |

MSc Classes: | Statistics LG.01, Thu 26 January, 9 February, 16 February, 2 March, 15-16 |

- Part C 2016 Q1
- Part C 2015 Q2 (a-c), Q3 (a-b)
- Part C 2013 Q3
- MSc 2012 Q6
- MSc 2014 Q7
- MSc 2016 Q7

- Problem Sheet 1 MSc students, Part B students. Solutions.
- Problem Sheet 2 all students, data and code. Solutions.
- Problem Sheet 3 all students. Solutions.
**[Note: the solution to problem 4 has been fixed.]** - Problem Sheet 4 all students Solutions [Note: Q5 uses notation that is not explained in the solutions; see Q1 here.]

- Lecture 1 (18/01):
**Introduction to machine learning, brief tour of topics covered in course (lecturer: Yee Whye Teh)**

slides - Lecture 2 (19/01):
**Empirical risk minimization, Bias/variance, Generalization, Overfitting, Crossvalidation (lecturer: Dino Sejdinovic)**

slides - Lecture 3 (25/01):
**Regularization and Logistic regression (Seth)**

slides, overfitting demo, regularization demo, logistic function demo [code] - Lecture 4 (26/01):
**Neural Networks (Seth)**

slides - Lecture 5 (1/02):
**Generative vs. Discriminative models, Maximum Likelihood Estimation, Mixture models (Mihaela)**

slides - Lecture 6 (2/02):
**Expectation Maximization (Mihaela)**

slides - Lecture 7 (8/02):
**Deep Learning (Seth)**

slides, deep learning demo from Google - Lecture 8 (9/02):
**Support Vector Machines (Seth)**

slides, read more about Lagrangian duality and the Karush-Kuhn-Tucker conditions in Chapter 5 of Boyd and Vandenberghe and read a different take on SVMs on pages 129-132 and 417-438 in ESL. - Lecture 9 (15/02):
**Clustering, K-means, Vector Quantization, Naive Bayes (Mihaela)**

slides - Lecture 10 (16/02):
**Nearest Neighbour Classification (Mihaela)**

slides - Lecture 11 (22/02):
**Support Vector Machines, LDA / QDA (Seth)**

slides - Lecture 12 (23/02):
**Training ML methods, Optimisation (Seth)**

Presentation was on whiteboard, based loosely on these notes. - Lecture 13 (1/03):
**Decision trees, CART (Mihaela)**

slides - Lecture 14 (2/03):
**Bagging decision trees, Random forests, and ROC curves (Dino)**

slides - Lecture 15 (8/03):
**Ensemble learning methods, Adaboost (Mihaela)**

slides - Lecture 16 (9/03):
**Reinforcement learning: a brief introduction (Mihaela)**

slides

- Installation
- RStudio
- Part A Statistical Programming at Oxford
- DataCamp tutorial
- Coursera R programming course
- intro to tidyverse (advanced)