Reasoning and Learning under Uncertainty

COMP64102 Reasoning and Learning under Uncertainty 15 Semester 2 Postgraduate Taught Level 6 Mauricio Alvarez Lopez Unit coordinator 1 FHEQ level (Framework for Higher Education Qualifications) ' Masters/Integrated Masters P4 ' European Credit Transfer & Accumulation System Rating : 7.5 <div class="OutlineElement Ltr SCXW139990802 BCX0" style="-webkit-tap-highlight-color:transparent;-webkit-text-stroke-width:0px;-webkit-user-drag:none;background-color:rgb(255, 255, 255);clear:both;color:rgb(0, 0, 0);cursor:text;direction:ltr;font-family:"Segoe UI", "Segoe UI Web", Arial, Verdana, sans-serif;font-size:12px;font-style:normal;font-variant-caps:normal;font-variant-ligatures:normal;font-weight:400;letter-spacing:normal;margin:0px;orphans:2;overflow:visible;padding:0px;position:relative;text-align:start;text-decoration-color:initial;text-decoration-style:initial;text-decoration-thickness:initial;text-indent:0px;text-transform:none;user-select:text;white-space:normal;widows:2;word-spacing:0px;">Machine learning is increasing being used for decision support in data driven applications. A key concept when making decisions based on predictive models is that of uncertainty, e.g., in applications of AI where safety or trustworthiness are required. Uncertainty quantification recognises that exact predictions are often out-of-reach due to theoretical or practical limitations.  This module studies different probabilistic machine learning models that incorporate uncertain reasoning and the mathematical concepts and algorithms required to learn such models from data.  </div> - introduce the main concepts behind reasoning and learning under uncertainty.- introduce the main machine learning models that are used for uncertainty quantification for different data structures.- provide practical experience of applying advanced probabilistic machine learning techniques to real-data problems.    <ol><li>Describe the fundamental concepts of uncertainty quantification.</li><li>Analyse the differences among statistical inference approaches.  </li><li>Explain the models and algorithms commonly used in probabilistic graphical models, state space models, Bayesian neural networks and Gaussian processes.</li><li>Apply an advanced uncertainty quantification model to a data-driven application using tools such as Scikit-learn and PyTorch, and probabilistic programs such as NumPyro and Stan. </li></ol> The syllabus for the Unit includes:1. Uncertainty quantification2. Statistical Inference3. Probabilistic graphical models4. State-space models5. Bayesian neural networks6. Gaussian processes  - Lecture (Week 1, 3, 5, 8, 10: 2-hr lecture, Week 2, 4, 7, 9, 11: 1-hr lecture; Week 6 is Reading Week)- Tutorials (Week 2, 4, 7, 9, 11: 1-hr tutorial solving theory exercises)- Computer Lab Sessions (Week 2, 4, 7, 9, 11: 2-hr lab session)    0 Other 50% 1 Written exam 50% Coding assignment 15 hrs 50%Exam 2 hrs 50%    <span style="-webkit-text-stroke-width:0px;display:inline !important;float:none;font-family:Arial, sans-serif;font-size:12px;font-style:normal;font-variant-caps:normal;font-variant-ligatures:normal;font-weight:400;letter-spacing:normal;orphans:2;text-align:-webkit-left;text-decoration-color:initial;text-decoration-style:initial;text-decoration-thickness:initial;text-indent:0px;text-transform:none;white-space:normal;widows:2;word-spacing:0px;">Coding assignment: Written solution provided 2 weeks after submission and feedback provided when marks are returned N - David Poole and Alan Mackworth, Artificial Intelligence: Foundations of Computational Agents,   Third Edition, Cambridge University Press, 2023- Kevin Murphy, Probabilistic Machine Learning: Advanced Topics, First edition, The MIT Press, 2023.- Osvaldo Martin, Ravin Kumar, Junpeng Lao, Bayesian Modeling and Computation in Python, CRC Press, 2022.- Christopher Bishop, Pattern Recognition and Machine Learning, Springer, 2007.    Lectures 15 Practical classes & workshops 10 Tutorials 5 0 120