COMSM0152 - Foundations of ProAI

Unit Information

This unit introduces students to a range of methodologies and applications across the field of AI. Each week will focus on a different theme, with three activities:

Required reading -- background reading to introduce the topic of the week
Guest seminars (TB2: Mondays, 1200-1300, Queen's 1.69) -- a guest speaker will present an introduction to their research area, connected to the week's topic
Group discussion (TB2: Fridays, 1100-1200, Queen's 1.69) -- student led discussion on the required reading and guest seminars. This is a chance to go over the important concepts and raise questions. The sessions will be chaired by a student, to facilitate the discussion (they are not required to present the topic). Link to spreadsheet with session chairs

This is a new unit, and you are a small cohort, so please do give us feedback at any time on how we can make the unit work better! In particular, if you find some topics assume prior knowledge you don't have, or would like to go further into certain topics, let us know. Please see below for some great all-round textbooks on AI, which may help you fill in some knowledge gaps... We also have a Team on MS Teams. Please do use it to ask us questions, discuss AI topics you are interested in, post interesting blogs, videos or papers, etc...

Staff

James Cussens (JC) Edwin Simpson (ES)

Unit Materials

TB1

TB2

Week	Reading (to complete before Monday's lecture)	Guest Seminar	Discussion chair (Link to spreadsheet with session chairs)	Lecturer responsible
1 (w/c 16/09/24)	Data and labels for image and Video understanding: ImageNet, the competition that kick-started the deep learning revolution in Computer Vision Kitti, a highly impactful dataset in vision and robotics. Optional extra: The winner of the ImageNet 2015 competition, ResNet.	Dima Damen -- tutorial on data and labels for video understanding	Jono	ES
2 (w/c 23/09/24)	Text embeddings and sequence processing: Deep contextualized word representations, Peters et al., 2018 Vector Semantics and Embeddings (intro to Section 6.4; 6.8 to end) RNNs and LSTMs (intro to Section 8.5)	Edwin Simpson	Moses	ES
3 (w/c 30/09/24)	Attention and Transformers. Both of the following are really good introductions to attention and transformers - you can pick one to read, or go through both if that helps your understanding. Please cover one of these before Monday: The Illustrated Transformer Jurafsky and Martin Chapter 9, "The Transformer" For the Friday reading group, please also read: Learning a Text-Video Embedding from Incomplete and Heterogeneous Data (and OPTIONALLY, for the keen) TOKEN MERGING: YOUR VIT BUT FASTER	Mike Wray	James	ES
4 (w/c 07/10/24)	Large language models: Jurafsky and Martin, Chapter 10 Jurafsky and Martin, Chapter 12, sections up to and inluding 12.4 Assessing the ability of LSTMs to learn syntax-specific dependencies, Linzen et al. (2016), TACL Beyond the limitations of any imaginable mecahnism: large language models and Psycholinguistics, Houghton, Kazanina and Sukumaran, Behavioral and Brain Sciences, (2023).	Conor Houghton -- linguistic perspectives on LLMs		ES
5 (w/c 14/10/24)	Philosophy of AI: An Analysis of the Interaction Between Intelligent Software Agents and Human Users Christopher Burr, Nello Cristianini and James Ladyman Language Models as Models of Language Raphaël Millière No Such Thing as a General Learner: Language models and their dual optimization Emmanuel Chemla and Ryan M. Nefdt	James Ladyman	Berenika	JC
6 (w/c 21/10/24)	CONSOLIDATION WEEK	(no seminar/reading group)
7 (w/c 28/10/24)	Knowledge representation and reasoning: R. Davis, H. Shrobe, and P. Szolovits. What is a Knowledge Representation? AI Magazine, 14(1):17-33, 1993 (16pp. + references) Delgrande, J. P., Glimm, B., Meyer, T., Truszczynski, M., and Wolter, F. (2023). Current and future challenges in knowledge representation and reasoning. arXiv preprint arXiv:2308.04161 (30pp. + appendices and an extensive bibliography) Additional reading: What are Non-classical Logics and Why Do We Need Them? An Extended Interview with Dov Gabbay and Leon van der Torre	Peter Flach	Moses	JC
8 (w/c 04/11/24)	Bayesian inference and decision making: An intuitive explanation of Bayes theorem (blog post) An introduction to variational autoencoders (Sections 1 and 2)	Laurence Aitchison (slides)	Jack	JC
9 (w/c 11/11/24)	Causality: Chapter 1 of Causal Inference in Statistics: A Primer by Pearl, Glymour and Jewell. You can skip Section 1.3 if you're already familiar with the material it contains. The tale wagged by the DAG: broadening the scope of causal inference and explanation for epidemiology by Krieger and Davey-Smith Comments on: The tale wagged by the DAG by Pearl Reply to Pearl: Algorithm of the truth vs real-world science by Krieger and Davey-Smith Supplementary reading: Schölkopf et al, Towards Causal Representation Learning	James Cussens		JC
10 (w/c 18/11/24)	Bias, fairness and transparency: European High-level expert group on AI, Ethics Guidelines for Trustworthy AI, read sections 4 and 5 of the assessment list on pp.28-30. Rebcca Burns, Artificial Intelligence is driving discrimination in the housing market, read up to the section "Not an excuse for lawbreaking behaviour." AI Fairness 360: read the short Credit Scoring tutorial via this link Miranda's slides	Miranda Mowbray	Jake	JC
11 (w/c 25/11/24)	Ethical and regulatory frameworks of AI: Charlesworth, A. Regulating Algorithmic Assemblages: Looking Beyond Corporatist AI Ethics, in Kohl, U. and Eisler, J. (eds.) (2022) Data-Driven Personalisation and the Law. Cambridge University Press. Charlesworth A. AI Regulation in the UK in Raposo, V.L. (ed.) The European Artificial Intelligence Act, Springer (forthcoming 2025). Additional Reading: Guihot, M., Matthew, A. F., and Suzor, N. P., (2020) Nudging Robots: Innovative Solutions to Regulate Artificial Intelligence, Vanderbilt Journal of Entertainment and Technology Law 20, 385-456. Additional Reading: EU Artificial Intelligence Act (Regulation (EU) 2024/1689) Chapter II - Prohibited AI Practices, Article 5 ; Chapter III- High Risk AI Systems, Arts. 6-17; Chapter V - General-Purpose AI Models, Arts. 51-56; Chapter X - Codes of Conduct & Guidelines, Art.95.	Andrew Charlesworth	Jack	JC
12 (w/c 09/12/24)	Assessment period	TB1 essay due
13 (w/c 13/01/25)	Discrete and continuous optimisation: Peter Bartlett. The Science of Modern Machine Learning: New Opportunities for Theory (link to talk sent privately) Bill Cook. The Traveling Salesman Problem: Postcards from the Edge of Impossibility	James Cussens	David	JC
14 (w/c 20/01/25)	Reinforcement learning: Chapter 1 of Sutton and Barto's book on reinforcement learning. It would be useful to read this before the Monday session as it gives an introduction to the topic. However, Taku is not assuming prior knowledge of RL in his talk. Playing Atari with Deep Reinforcement Learning , a classic deep RL paper Q-learning Decision Transformer: Leveraging Dynamic Programming for Conditional Sequence Modelling in Offline RL by Yamagata, Khalil, and Santos-Rodriguez	Taku Yamagata	Berenika	ES
15 (w/c 27/01/25)	Tactile robotics: Nathan has kindly provided us a sneak preview of two of his papers: you can find them here on Blackboard (you'll need to log in first). Please don't share the papers as they have not yet been published.	Nathan Lepora	Jono	ES
16 (w/c 03/02/25)	Multi-agent systems: Please skim through this chapter: Gerhard Weiss (2013). 2013. Intelligent agents. In Multiagent Systems, 2nd ed., pages 3--50. MIT Press. [eBook available through University of Bristol library] Pradeep K. Murukannaiah, Nirav Ajmeri, Catholijn Jonker, and Munindar P. Singh. New Foundations of Ethical Multiagent Systems. Proceedings of the 19th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), Blue Sky Idea Track, Auckland, May 2020, pages 1706--1710. Jessica Woodgate, Paul Marshall, and Nirav Ajmeri. Operationalising Rawlsian Ethics for Fairness in Norm-Learning Agents. Proceedings of the 39th AAAI Conference on Artificial Intelligence (AAAI), Philadelphia, Feb 2025, pages 1--9.	Nirav Ajmeri	Bezawit	JC
17 (w/c 10/02/25)	Weak supervision: An introduction to the topic: Snorkel AI webpages/videos Snorkel: rapid training data creation with weak supervision (2017). Learning from Crowds (Raykar et al., 2010) – a canonical paper on treating crowdsourced labels as weak supervision	Raul Santos-Rodriguez	James	ES
18 (w/c 17/02/24)	CONSOLIDATION WEEK	No seminar/reading group
19 (w/c 24/02/25)	Learning from Temporal Data for Healthcare: Have a look at this central resource for time series classification datasets and methods The great time series classification bake off Deep learning for time series classification: a review Supplementary reading: EEG2Rep: Enhancing Self-supervised EEG Representation Through Informative Masked Inputs Supplementary reading: ROCKET: exceptionally fast and accurate time series classification using random convolutional kernels Supplementary reading: The great multivariate time series classification bake off	Zahraa Abdallah	Berenika	JC
20 (w/c 03/03/25)	Explainable and interpretable AI: Counterfactual explanations and how to find them: literature review and benchmarking If Only We Had Better Counterfactual Explanations: Five Key Deficits to Rectify in the Evaluation of Counterfactual XAI Techniques Explainable AI for Time Series Classification: A Review, Taxonomy and Research Directions Supplementary reading: EEG2Rep: Counterfactual Explanations for Machine Learning: A Review	Weiru Liu	Jake	ES
21 (w/c 10/03/25)	Privacy: The AI & data protection risk toolkit from the UK Information Commissioner's Office, focussing on column I, Practical Steps to Reduce the Risk. (Select the "Risk toolkit" tab.) Steps 5 and 6 of How do we do a DPIA? This Wes Davis article	Miranda Mowbray	Jack	JC
22 (w/c 17/03/25)	MLOps and deploying AI in production: Resilience-aware MLOps for AI-based medical diagnostic system (2024) Machine Learning Operations (MLOps): Overview, Definition, and Architecture (2023) Have a browse of this (a huge resource where you can find information about any MLOps topic of interest: Awesome MLOps Github Repo Extra: Scaling AI responsibly: Leveraging MLOps for sustainable machine learning deployments (2024)	Jose Navarro, Cleo AI (external guest speaker)	Bezawit	ES
23 (w/c 24/03/25)	Human-in-the-loop AI -- design and evaluation: Cathy O’Neil, C. (2016) Weapons of Math Destruction. Penguin: Introduction, Chapter 1. Crawford, K. (2021) Atlas of AI. Yale: Chapter 2 Labor, Chapter 4 Classification. Ghosh, P. L. Posner, K. Hyland, S. Van Cleve W, Bristow M, Long DR, Palla K, Nair B, Fong, C., Pauldine, R, Vavilala, M.and Kenton O'Hara.K (2023) Framing Machine Learning Opportunities for Hypotension Prediction in Perioperative Care: A Socio-technical Perspective. ACM Transactions on Computer-Human Interaction, 30(5) Palla K, Hyland SL, Posner K, Ghosh P, Nair B, Bristow M, Paleva Y, Williams B, Fong C, Van Cleve W, Long DR, Pauldine R, O'Hara K, Takeda K, Vavilala MS. (2022) Intraoperative prediction of post anaesthesia care unit hypotension. Br J Anaesth,128(4):623-635.	Kenton O'Hara	James	ES

Assessment Details

Deadline for submission of TB2 essays is: 28 April 2025

After each Teaching Block students submit an essay of about 5,000 words (10 pages) on a research topic jointly chosen by them and their Academic Mentor. The essay should describe the background, state of the art, and open challenges with regard to the chosen topic. Each essay is assessed on a pass/fail basis in terms of scholarly content and academic writing. Narrative feedback is also provided, indicating strong points as well as areas for improvement. Passing the unit requires passing both essays. More guidance will be provided by the unit lecturers during the first term.

Text books

Bishop, C. M., Pattern recognition and machine learning (2006). This is one of the best ML textbooks and will provide a solid foundation across many aspects of ML. The book is freely available here.
Russell, S. and Norvig, P., Artificial Intelligence, A Modern Approach, 4th Edition (2020). The canonical introduction to AI, 3rd edition available at here.
Jurafsky, D. and Martin, J.H., Speech and Language Processing, 3rd edition drafts (2024). A great NLP textbook, but very readable and an interesting way to think about the challenges of designing AI systems in general. Also good for showing a contrast between deep learning and feature engineering approaches. Online only, here.
Murphy, K., Probabilistic Machine Learning: An Introduction (2022) and Murphy, K., Probabilistic Machine Learning: Advanced Topics (2023). A more recent ML textbook with particularly good coverage of probabilistic methods, freely available via here.

Some of you may prefer videos to books. Youtube has many video lectures, including the Machine Learning course from Stanford by Andrew Ng, which is very good.