Postgraduate Course: Online Learning and Decision Making (CMSE11642)
Course Outline
| School | Business School |
College | College of Arts, Humanities and Social Sciences |
| Credit level (Normal year taken) | SCQF Level 11 (Postgraduate) |
| Course type | Online Distance Learning |
Availability | Not available to visiting students |
| SCQF Credits | 10 |
ECTS Credits | 5 |
| Summary | This course examines the intersection of online learning and data-driven decision-making. You'll gain a good understanding of online learning framework, then explore strategies for choosing the best decision when faced with limited information what is known as the multi-armed bandit problem. You will learn theories to balance exploration with exploitation, minimise regret, and confidently identify effective solutions under limited information The course will also examine real-world examples, for example, model selection in AI, clinical trial design in healthcare, choice modelling in revenue management. |
| Course description |
This advanced course provides a rigorous exploration of the theoretical foundations and practical methodologies underpinning online learning and data-driven decision-making processes. The course content is structured to build incrementally, starting with the foundational concepts and progressively integrating more complex theories and applications. Through this course, students will gain proficiency in designing, analysing, and implementing online decision-making algorithms.
Outline Content
- Stochastic Dynamic Programming: Master the modeling and solution of sequential decision problems. Develop fluency in Markov Decision Processes, the Bellman Equation, and techniques like value iteration and policy iteration.
- Multi-armed Bandit: Learn about algorithms and strategies to effectively handle the exploration-exploitation trade-off. Delve into methods like upper confidence bound, Thompson sampling and knowledge gradient.
- Applications in Online Decision Making: Investigate real-world scenarios across industries. Analyse how online decision making frameworks lead to better outcomes.
The syllabus is adaptable to accommodate the latest developments in the field.
Student Learning Experience
This course blends theory and practice for a hands-on learning experience. Expect a mix of lectures introducing core concepts, followed by workshops where you'll apply those concepts using Python (or a similar language). You'll demonstrate your understanding through individual assignments showcasing your ability to model and solve online decision-making challenges.
|
Entry Requirements (not applicable to Visiting Students)
| Pre-requisites |
|
Co-requisites | |
| Prohibited Combinations | |
Other requirements | None |
Course Delivery Information
|
| Academic year 2024/25, Not available to visiting students (SS1)
|
Quota: None |
| Course Start |
Block 1 (Sem 1) |
| Course Start Date |
16/09/2024 |
Timetable |
Timetable |
| Learning and Teaching activities (Further Info) |
Total Hours:
100
(
Lecture Hours 10,
Seminar/Tutorial Hours 2,
Programme Level Learning and Teaching Hours 2,
Directed Learning and Independent Learning Hours
86 )
|
| Assessment (Further Info) |
Written Exam
0 %,
Coursework
100 %,
Practical Exam
0 %
|
| Additional Information (Assessment) |
100% Essay (Individual) - 1,500 words - Assesses all course Learning Outcomes |
| Feedback |
Formative: Feedback will be provided throughout the course.
Summative: Feedback will be provided on assessments within agreed deadlines. |
| No Exam Information |
Learning Outcomes
On completion of this course, the student will be able to:
- Critically evaluate and formulate sequential decision-making problems using stochastic dynamic programming models.
- Analyse the trade-offs between exploration and exploitation in multi-armed bandit problems, and apply appropriate algorithms for informed decision-making.
- Design and implement online learning strategies for various decision-making scenarios within real-world contexts.
- Communicate decision-making processes and outcomes effectively with both technical and non-technical stakeholders.
|
Reading List
Optimal Learning , ISBN : 9780470596692
Bandit Algorithms, ISBN: 9781108571401
Discrete Choice Modelling and Air Travel Demand : Theory and Applications, ISBN : 9780754681267 |
Additional Information
| Graduate Attributes and Skills |
Communication, ICT, and Numeracy Skills
After completing this course, students should be able to:
Critically evaluate and present digital and other sources, research methods, data and information; discern
their limitations, accuracy, validity, reliability and suitability; and apply responsibly in a wide variety of
organisational contexts.
Knowledge and Understanding
After completing this course, students should be able to:
Demonstrate a thorough knowledge and understanding of contemporary organisational disciplines; comprehend the role of business within the contemporary world; and critically evaluate and synthesise primary and secondary research and sources of evidence in order to make, and present, well informed and transparent organisation-related decisions, which have a positive global impact.
Identify, define and analyse theoretical and applied business and management problems, and develop approaches, informed by an understanding of appropriate quantitative and/or qualitative techniques, to explore and solve them responsibly.
|
| Keywords | Online Learning,Data-driven Decision Making |
Contacts
| Course organiser | Dr Xin Fei
Tel: (0131 6)50 8074
Email: Xin.Fei@ed.ac.uk |
Course secretary | Ms Connie Wong
Tel:
Email: ywong@ed.ac.uk |
|
|
University Homepage