Dr Tengyao Wang

This course is available on the MPA in Data Science for Public Policy, MSc in Applied Social Data Science, MSc in Data Science, MSc in Econometrics and Mathematical Economics, MSc in Geographic Data Science, MSc in Health Data Science, MSc in Management of Information Systems and Digital Innovation, MSc in Quantitative Methods for Risk Management, MSc in Statistics, MSc in Statistics (Financial Statistics), MSc in Statistics (Financial Statistics) (Research), MSc in Statistics (Research), MSc in Statistics (Social Statistics) and MSc in Statistics (Social Statistics) (Research). This course is available with permission as an outside option to students on other programmes where regulations permit.

This course has a limited number of places (it is controlled access) and demand is typically high. This may mean that you are not able to get a place on this course. MSc in Data Science students are given priority for enrollment in this course. 

The course provides a broad overview on fundamental concepts and algorithms of artificial intelligence systems, with focus on search methods, knowledge representation, game playing, logical and probabilistic reasoning, supervised learning and reinforcement learning. We use state-of-the-art data science and artificial intelligence Python libraries and tools to translate the studied principles and methods into practice, and to gain hands-on experience in data analysis.

• Introduction: aims, history, rational actions, and agents.
• Simple uninformed search methods: graph search, tree-like search, best-first search, breadth-first search, uniform search, depth-first search, limited depth-1st search, iterative deepening search.
• Advanced informed search methods: more sophisticated heuristic search algorithms, A* search, local search, hill-climbing search, simulated annealing, local beam search, genetic algorithm, conditional plan, AND-OR search, belief states.
• Game playing: adversarial search, the minmax algorithm and its shortcomings, improving minimax using alpha-beta pruning, Type A (wide) and Type B (deep) strategies, stochastic games, EXPECTIMAX search.
• Constrained satisfaction problems (CSPs): standardising search problems to a common format, backtracking algorithm for CSPs, heuristics for improving the search for a solution, constraint propagation and consistency, solving Sudoku.
• Knowledge representation and logical reasoning: representation of common sense knowledge, inference and knowledge representation schemes, propositional logic, syntax, semantics and entailment.
• Probabilistic reasoning: representing knowledge in uncertain domain, graphical models, Bayesian networks, statistical inference in Bayesian networks.
• Supervised learning: learning from examples, hypothesis space, loss and risk, model selection, regularization, linear regression and classification, logistic regression, kernel machines, multilayer perceptron and the backpropagation algorithm.
• Reinforcement learning: reinforcement learning problem formulation by using Markov Decision Processes, dynamic programming, Bellman optimality solution, simple tabular solution methods.

20 hours of lectures and 15 hours of classes in the AT.

This course will be delivered through a combination of lectures and classes totalling a minimum of 35 hours across Autumn Term. This course includes a reading week in Week 6 of Autumn Term.

Students will be expected to produce 10 problem sets in the AT.

• Stuart Russell, Peter Norvig. Artificial Intelligence: A Modern Approach. 4th edition, Pearson, 2020. [http://aima.cs.berkeley.edu/]
• David Poole, Alan Mackworth. Artificial Intelligence: Foundations of Computational Agents, 2nd Edition, Cambridge University Press, 2017. [https://artint.info/]
• Kevin Murphy. Probabilistic Machine Learning. 2021-2022. [https://probml.github.io/pml-book/]
• Christopher M. Bishop. Pattern Recognition and Machine Learning. Springer-Verlag, 2007. [https://www.microsoft.com/en-us/research/people/cmbishop/prml-book/]
• Aston Zhang, Zachary C. Lipton, Mu Li, and Alexander J. Smola. Dive into Deep Learning, 2021. [http://d2l.ai/]
• Ian Goodfellow, Yoshua Bengio and Aaron Courville. Deep Learning, MIT Press, 2016. [https://www.deeplearningbook.org/]
• Richard S. Sutton, Andrew G. Barto. Reinforcement Learning: An Introduction. 2nd edition, MIT Press, 2018. [http://incompleteideas.net/book/the-book.html]
• TensorFlow, An Open Source Software Library for Machine Intelligence. [http://www.tensorflow.org]
• Jake VanderPlas. Python Data Science Handbook. O'Reilly Media, Inc., 2017. [https://www.oreilly.com/library/view/python-data-science/9781491912126/ , https://jakevdp.github.io/PythonDataScienceHandbook/]
• Mark Lutz. Learning Python, 5th Edition. O'Reilly Media, Inc., 2013. [https://www.oreilly.com/library/view/learning-python-5th/9781449355722/]

Project (80%) in the AT.
Continuous assessment (20%) in the AT Week 8.

The problem set submitted by students is assessed (20% in total). In addition, there is a graded take-home research project (80%) which is completed by students in groups, in which they demonstrate the ability to apply and train an appropriate model to a specific problem and dataset using principles they have learnt in the course.