## ANU COMP 4670 / 8600 - Statistical Machine Learning

A broad but thorough intermediate level course of statistical machine learning, emphasising the mathematical, statistical, and computational aspects

## Overview

Statistical Machine Learning plays a key role in science and technology. Some of the basic questions raised are:

What is a good model for the available data?

How can we fit the parameters of the model to the available data?

How will a model perform on data which has yet to be observed?

This course provides a broad but thorough intermediate-level study of the methods and practices of statistical machine learning, emphasising the mathematical, statistical, and computational aspects. Students will learn how to implement efficient machine-learning algorithms on a computer based on principled mathematical foundations. Topics covered will include Bayesian inference and maximum likelihood modelling; regression, classification, density estimation, clustering, principal and independent component analysis; parametric, semi-parametric, and non-parametric models; basis functions, neural networks, kernel methods, and graphical models; deterministic and stochastic optimisation; overfitting, regularisation, and validation.

The course will use Python 3 and Jupyter notebook for all tutorials and assignment/exam questions involving programming.

## Course schedules

### Timetable

### Week 1A (Feb 21)

Course Introduction

### Week 1B (Feb 22)

Machine Learning 101,

Probabilities, Model Selection

### Self-Assessment

Assignment 0 - self-assessment

### Self-Assessment

Jupyter Notebook, Matrices

### Week 2A (Feb 28)

Linear Regression

### Week 2B (Feb 29)

Bayesian Linear Regression

### Assignment

Assignment 1 released

Video Assignment released

### Week 2 Reading

Pre-read: Gaussian Mixture Model, Mathematics for Machine Learning, Chapter 11

### Week 2 Tutorial

Linear Algebra, Optimisation, Probabilities

### Week 3A (March 7)

Linear Classification

### Week 3B (March 8)

Expectation Maximisation,

Mixture Models

### Week 3 Reading

Pre-read: Kmeans,

Pattern Recognition and Machine Learning, Chapter 9.1

### Week 3 Tutorial

Regression

### Week 4A (March 14)

Generalisation

### Week 4B (March 15)

Neural Networks

### Quiz 1

### Week 4 Reading

Pre-read: Patterns, Predictions and Actions: A Story about Machine Learning, Chapter 5

### Week 4 Tutorial

Classification

### Week 5A (March 21)

Automatic Differentiation

and its application to Neural Networks

### Week 5B (March 22)

Linear and Non-linear Component Analysis, including Auto Encoder, Recommender System 101

### Week 5 Reading

Pre-read: Principal Component Analysis, Mathematics for Machine Learning, Chapter 10

### Week 5 Tutorial

Expectation Maximisation

### Week 6A (March 28)

Kernel Methods

### Week 6B (March 29)

Kernel Machines

### Assignment

Assignment 1 due [Mon 12:00 noon]

### Week 6 Tutorial

Neural Networks

### Semester Break (Apr 3 - Apr 16)

### Assignment

Assignment 2 released

### Week 7 Tutorial

Dimension Reduction

### Quiz 2

### Week 8 Tutorial

Kernels

### Week 9 Tutorial

Gaussian Process

### Week 10 Tutorial

Sampling

### Assignment

Assignment 2 due

[Monday 12:00 noon]

### Week 11 Tutorial

Graphical Models

### Assignment

Video Assignment due

[Monday 12:00 noon]

Labs and tutorials: refer to mytimetable

### Exam Timetable

- Final exam: June 2, 2023, 9:50am-12:50pm

## Course staff

### Lecturers

School of Computing

School of Computing /

Research School of Astronomy and Astrophysics

### Tutors

Chamin Hewa KONEPUTUGODAGE

(Head Tutor)

Alexander SOEN

(Head Tutor)

Anupama ARUKGODA

Dillon CHEN

Ziyu CHEN

Lydia LUCCHESI

Samuel JOLLEY

Buddhi KOTHALAWALA

Shidi LI

James Yuanchu LIANG

Evan MARKOU

Lachlan McGINNES

Josh NGUYEN

Vimukthini PINTO

Taylor Zishan QIN

Belona SONNA

Chunyi SUN

Tianyu WANG

Zhifeng WANG

Hansheng XUE

Allen Qinyu ZHAO

Haiqing ZHU

## Textbooks

Christopher M. Bishop:

Pattern Recognition and Machine Learning

Springer, 2006 (selected parts)

We also recommend

Deisenroth, Faisal, and Ong, Mathematics for Machine Learning. Cambridge University Press.

Moritz Hardt and Benjamin Recht, Patterns, Predictions and Actions: A Story about Machine Learning

MacKay, Information Theory, Inference, and Learning Algorithms, Cambridge University Press

Murphy, Probabilistic Machine Learning: An Introduction, MIT Press, 2021

## Course sites

All read-only content will be on the course web page -- this page!

The lectures and most tutorials will be in person. Microsoft teams (ANU edition) will be used to stream lectures via video (in addition to echo360) and host online tutorials/labs. Lectures will be recorded, but online tutorials will not be recorded.

EdSTEM is a new platform replacing piazza (with a few more functionalities)

Wattle will be used for exams, quizzes and surveys. SML 2023 wattle site can be found here.

A brief class FAQ can be found here.

## Assessments

Quizzes (2% x 2)

Assignments (18% x 2)

Video assignment (20%)

Final exam (40%)

Online quiz expectations

The quiz will be conducted on Wattle. It will be automatically graded with answers released after the closing date.

Students can attempt the quiz once, with no time limit.

Open book -- students are expected to complete the quiz by themselves and are free to consult the textbook, notes, or relevant internet resources.

The quiz will be redeemable with the final exam, i.e. score for each quiz is calculated as Qx' = max(Qx, Final), where Qx is the raw quiz score for Quiz 1 (Q1) or Quiz 2 (Q2), out of 100. Final is the score for the final exam out of 100.

There will be no late period for either quiz. Special consideration requests will also not be accepted due to the rapid feedback cycle and redeemable nature of the quizzes.

## Assignment

Assignments

Assignments 1 and 2 are individual assignments with conceptual, mathematical, and programming components. Submission instructions will be made available closer to time.

Video assignment

The video assignment is an individual assignment.

Each student is expected to upload a video discussing one topic from the assignments or labs and the thinking behind it.

The length of the video should be between 4 to 8 minutes, with an under- and over-length penalty being 1 point per 10 seconds (or part thereof).

The grading scheme for the video assignment will be made available in advance of the due date.

Late policy

This policy applies to Assignment 1, Assignment 2, and the video assignment.

Assignment submissions that are late from 1 min to 24 hours attract a 5% penalty (of possible marks available).

Submissions late by more than 24 hours will not be accepted.

## Enrollment

To enrol in this course, you must have completed the prerequisites as per the COMP4670 or COMP8600 course description.

The topics covered in this course overlap with several courses in the major of Statistical Data Analytics. Please look at the first few tutorial sheets for an indication of the kinds of mathematics and statistics we will build upon.

Other enrollment info, including obtaining permission codes, is covered in the FAQ.