COMS30117 - Deep Learning
Unit Information
Welcome to COMS30117. 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 - 03/08/2026 Under Construction for 26/27!
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 (tbc).Staff
| Michael Wray (MW) | Unit Director |
| Tilo Burghardt (TB) |
Teaching Assistants
Jiahe Zhou (JZ), Maryam Heidari Kahrangini (MK), Omar Emara (OE), Prajwal Gatti (PG), Rhodri Guerrier (RG), Sam Pollard (SP), Siddhant Bansal (SB), Tomos Sherlock (TS)
Unit Materials
| Wks | Tuesday 15:00-18:00 | Friday 09:00-11:00 | Labs |
| 1 |
22/09/2026 - 15:00 - Chemistry LT3 LECTURE 1 INTRODUCTION TO THE UNIT intro slides BASICS OF ARTIFICIAL NEURAL NETWORKS Introduction, Neural Networks, Perceptron, Cost Functions, Gradient Descent, Delta Rule, Deep Networks PDF Slides, Recording |
25/09/2026 - 09:00 - Chemistry Building LT3 LECTURE 2 TOWARDS TRAINING DEEP FORWARD NETWORKS Network Representation, Computational Graphs, Reverse Auto-Differentiation PDF Slides, Extra Recap Recording Lecture 2 Refresher (first part of video) |
GETTING STARTED: RECAP WORKSHEETS: Lab 0 - Python (Homework) |
| 2 |
29/09/2026 - 15:00 - MVB 2.11 PRACTICAL 1 YOUR FIRST FULLY CONNECTED LAYER Fully Connected Layers Stochastic Gradient Descent Slides |
02/10/2026 - 09:00 - Chemistry Building LT3 LECTURE 3 BACKPROPAGATION ALGORITHM The Backpropagation Algorithm in Full Detail, Activation Functions PDF Slides Extra Recap Recording (second part of video) LECTURE 4 OPTIMISATION TECHNIQUES Stochastic Gradient Descent, Nesterov Momentum, RMSProp, Newton's Method, AdaGrad, Adam, Saddle Points PDF Slides Extra Recap Recording |
29/09/2026, (MVB 2.11) - 3hrs Lab 1 - Training your first Deep Neural Network |
| 3 |
06/10/2026 - 15:00 - MVB 2.11 PRACTICAL 2 YOUR FIRST CONVOLUTIONAL CONNECTED LAYER Convolutional Layers, Pooling Slides |
LECTURE 5 CONVOLUTIONAL NEURAL NETWORKS sharing parameters, conv layers, pooling, CNN architectures Slides |
06/10/2026 (MVB 2.11) - 3hrs Lab 2 - Your First Convolutional Connected Network |
| 4 |
13/10/2026 - 15:00 - MVB 2.11
PRACTICAL 3 Hyperparameters Error rate monitoring (training/validation/testing) Batch-based training Learning rate Weight Freezing Batch normalisation Parameter intialisation Slides |
09/10/2026 - 09:00 - Chemistry Building LT3 LECTURE 6 COST FUNCTIONS, REGULARISATION AND DEPTH SoftMax, Cross Entropy, L1 and L2 Regularisation, DropOut, DropConnect, Depth Considerations PDF Slides Extra Recap Recording |
13/10/2026, (MVB 2.11) - 3hrs Lab 3 - Hyperparameters |
| 5 |
20/10/2026 - 15:00 - MVB 2.11 PRACTICAL 4 Data Augmentation Debugging strategies Dropout Slides |
16/10/2026 - 09:00 - Chemistry Building LT3 Mid-Term Support Session |
20/10/2026, (MVB 2.11) - 3hrs Lab 4 - Data Augmentation |
| 6 | READING WEEK - Mid Term for ALL MAJOR unit students 30/10/2026 - MVB - 2.11 - 10:00-11:00 | ||
| 7 |
03/11/2026 - 15:00 - MVB 2.11 Continuation Lab |
16/10/2026 - 09:00 - Chemistry Building LT3 LECTURE 7 RECURRENT and RELATIONAL NEURAL NETWORKS RNN, encoder-decoder, Transformers Slides |
10/11/2026, (MVB 2.11) - 3hrs Catch-Up |
| 8 |
10/11/2026 - 15:00 - MVB 2.11
PRACTICAL 5 Transformers Transformer Encoders Slides |
23/10/2026 - 09:00 - Chemistry Building LT3 LECTURE 8 GENERATIVE MODELS Autoregressive models Slides |
03/11/2026, (MVB 2.11) - 3hrs Lab 5 - Transformers |
| 9 | 17/11/2026, 15:00 [2 hours], (MVB 2.11) CW Support Session | - | - |
| 10 | 24/11/2026, 15:00 [2 hours], (MVB 2.11) CW Support Session | - | - |
| 11 | 01/12/2026, 15:00 [2 hours], (MVB 2.11) CW Support Session | - | - |
| 12 |
08/12/2026, MVB 2.11 15:00 Exam Support Session |
- | - |
| 13 | DECEMBER EXAMS - Final for MINOR unit students | ||
Assessment Details
- Coursework (for Major option): You are requested to re-implement work to be specified, and provide your code as well as a final report. Coursework completed in groups (up to 4)
- Mid-term test (for Major option): The theoretical components of the unit up to week 5 are examinable. No code would be written in the exam, but code can be provided to answer questions. Calculators are allowed.
- Exam (for Minor option): Both the theoretical and practical components of the unit are examinable. No code would be written in the exam, but code can be provided to answer questions. Calculators are allowed.
Assessment Details - Coursework
The coursework will be released during TB1
Assessment Details - Exam
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 COMS30117 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