UNIT INFO

COMSM0045 - Applied Deep Learning

ADL Banner

Unit Information

Welcome to COMSM0045. The unit introduces the students to deep architectures for learning linear and non-linear transformations of big data towards tasks such as classification and regression. The unit paves the path from understanding the fundamentals of convolutional and recurrent neural networks through to training and optimisation as well as evaluation of learnt outcomes. The unit's approach is hands-on, focusing on the 'how-to' while covering the basic theoretical foundations. For further general information, see the syllabus for the unit.

UPDATE - 24/06/2025 Currently Under Construction for 25/26! PLEASE NOTE: lecture content will be updated, slides below are placeholder and may change until the lecture If you have any questions, head to the unit teams.

Staff

Michael Wray (MW)Unit Director
Tilo Burghardt (TB)

Teaching Assistants

Omar Emara (OE), Prajwal Gatti (PG), Rhodri Guerrier (RG), Sam Pollard (SP), Saptarshi Sinha (SS), Siddhant Bansal (SB), Yini Li (YL)


Unit Materials

Wks Monday 16:00-18:00 Tuesday 15:00-18:00 Labs
1 22/09/2025 - 16:00 - Queens BLDG 1.07
Wk1 - LECTURE 1
INTRODUCTION TO THE UNIT
intro PDF

BASICS OF ARTIFICIAL NEURAL NETWORKS
(Queens Building 1.07, in-person)
(Introduction, Neural Networks, Perceptron, Cost Functions, Gradient Descent, Delta Rule, Deep Networks)
PDF Slides, Recording
23/09/2025 - 15:00 - Queens BLDG 1.07
Wk1 - LECTURE 2
TOWARDS TRAINING DEEP FORWARD NETWORKS
(MVB 1.15, in-person + lecture recap)
(Network Representation, Computational Graphs, Reverse Auto-Differentiation)
PDF Slides, Extra Recap Recording Lecture 2 Refresher (first part of video)
GETTING STARTED:

Register Individually on BlueCrystal4
(details see below)

RECAP WORKSHEETS:
-Lab0 - Python (Homework)
2 29/09/2025 - 16:00 - Queens BLDG 1.07
Wk2 - LECTURE 3
BACKPROPAGATION ALGORITHM
(Queens Building 1.07, in-person + recorded lecture)
(The Backpropagation Algorithm in Full Detail, Activation Functions)
PDF Slides
Extra Recap Recording (second part of video)
Wk2 - LECTURE 4
OPTIMISATION TECHNIQUES
(Queens Building 1.07, in-person + recorded lecture)
(Stochastic Gradient Descent, Nesterov Momentum, RMSProp, Newton's Method, AdaGrad, Adam, Saddle Points)
PDF Slides
Extra Recap Recording
30/09/2025 - 15:00 - MVB 1.15 -
PRACTICAL 1
Your first fully connected layer
gradient descent
stochastic gradient descent
Slides
30/09/2025, (MVB 1.15) - 3hrs
-BC4 Setup
Lab 1 - Training your first Deep Neural Network
3 06/10/2025 - 16:00 - Queens BLDG 1.07
Wk3 - LECTURE 5
COST FUNCTIONS, REGULARISATION AND DEPTH
(Queens Building 1.07, in-person + recorded lecture)
(SoftMax, Cross Entropy, L1 and L2 Regularisation, DropOut, DropConnect, Depth Considerations)
PDF Slides
Extra Recap Recording
07/10/2025 - 15:00 - MVB 1.15 -
PRACTICAL 2
Your first convolutional connected layer
Slides
07/10/2025, (MVB 1.15) - 3hrs
Lab 2 - Your First Convolutional Connected Network
Wk3 - LECTURE 6
CONVOLUTIONAL NEURAL NETWORKS
(Queens Building 1.07, in-person + recorded lecture)
(sharing parameters, conv layers, pooling, CNN architectures)
Slides
Recording
4 13/10/2025 - 16:00 - Queens BLDG 1.07
Wk4 - LECTURE 7
RECURRENT and RELATIONAL NEURAL NETWORKS
(Queens Building 1.07, in-person + Recorded)
(RNN, encoder-decoder, Transformers)
Slides
14/10/2025 - 15:00 - MVB 1.15 -
PRACTICAL 3
Error rate monitoring (training/validation/testing)
Batch-based training
Learning rate
Weight Freezing
Batch normalisation
Parameter intialisation
Slides
14/10/2025, (MVB 1.15) - 3hrs

Lab 3 - Hyperparameters
5 20/10/2025 - 16:00 - Queens BLDG 1.07
Wk5 - LECTURE 8
GENERATIVE MODELS
(Queens Building 1.07, in-person + Recorded)
(Autoregressive models)
Slides
21/10/2025 - 15:00 - MVB 1.15 -
PRACTICAL 4
Data Augmentation
Debugging strategies
Dropout
Slides
21/10/2025, (MVB 1.15) - 3hrs

Lab 4 - Data Augmentation
6 READING WEEK - Mid Term for MAJOR unit students 30/10/2025 - MVB - 1.07 - 13:00-14:00
7 - 04/11/2025 - 15:00 - MVB 1.15 -
PRACTICAL 5
Transformers
Slides
04/11/2025, (MVB 1.15) - 3hrs
Lab 5 - Transformers
8 - 11/11/2025 - 15:00 - MVB 1.15 -
Continuation Lab
11/11/2025, (MVB 1.15) - 3hrs

9 - 18/11/2025, 15:00 [2 hours], (MVB 1.15) -
CW Support Session
-
10 - 25/11/2025, 15:00 [2 hours], (MVB 1.15) -
CW Support Session
-
11 - 02/12/2025, 15:00 [2 hours], (MVB 1.15) -
CW Support Session
-
12 - 09/12/2025, Queens BLDG 1.07 -
Exam Support Session
-
13 DECEMBER EXAMS - Final for MINOR unit students

Assessment Details


Assessment Details - Coursework

The coursework will be released during TB1


Assessment Details - Exam

More details regarding the exam and in-class test coming soon. You can find previous papers here, but please note that these were from when the unit only contained one 2 hour exam.

Please note that you cannot take notes into the exam (it is closed book), but calculators are permitted.


Github

All technical resources will be posted on the COMSM0045 ADL Github organisation. If you find any issues, please kindly raise an issue in the respective repository.


Textbook

Recommended Reading:Simon J.D (2023). Prince. Understanding Deep Learning, MIT Press

You can also check out, written pre transformers, the older course book which we still recommend: Goodfellow et al (2016). Deep Learning. MIT Press