0x010088F6312065A70A44B630A6848498EAD80093CFEAFE161DC94D83C7C5C9A5A748B8

This course is part of a portfolio of four Ph.D. courses developed under a research project funded by The Research Council of Norway, COMPAMA: https://www.ntnu.edu/compama/courses. These courses are designed to be given by an external lecturer (preferably an expert in the topic) in one teaching week. They are open to NTNU students and students from other universities in Norway and abroad. Each course should run twice. In this case, Spring 2022 and Spring 2024. Given that this was the first edition, there were no action plans from before.

The external lecturer gave the course in a way that theoretical lectures were merged with implementation exercises. The learning outcome in the course are: -	explain and implement the techniques learned;-	choose the more suitable approach for a specific economic application;-	recognize the opportunities and challenges of using AI in each context.In this sense, the type of lecturing was very suitable to achieve these goals.The students appreciated this mixed type of lecture. However, some students said that they would prefer to have the opportunity to connect online and have recordings instead of spending the time coming to the campus. Besides needing approval from the external lecturer to do it, in our opinion this wipes off one of the main goals of the COMPAMA project, which is to create new research networks. Inviting a specialist from abroad to teach at NTNU and then having no students in the room would be embarrassing and not meaningful. Furthermore, some students felt that some theoretical parts, such as proofs of results, were not so interesting to them. Also, they would like to see other applications beyond Finance. This is something we will take into consideration for the next edition.The assessment of the course is defined by the responsible for the course at NTNU and the external lecturer. In this specific course, we have defined to have a project to assess the knowledge assimilated by the students. The students worked in a group. Their goal was to solve a specific problem using reinforcement learning. They needed to understand the problem, implement the method and discuss the results in a report. All these requests are in line with the learning outcomes of the course. The students were in general satisfied with it but they would prefer to have a description of it in advance (to be able to use the lectures to clarify points about it). Some said that the project could be smaller and others would prefer an oral exam instead. For the next edition, we will ask the lecturer to define the form of assessment before the course start-up. Then, the students can prepare themselves in advance. Regarding the type of assessments, being a course that is important to be able to explain and implement different techniques, it seems that a project is still more suitable than an oral exam. Given the format of the course, we did not have a reference group. Instead, at the end of the course, an anonymous survey was shared with the students. Two students answered it, from a total of three students. The feedback that I have shared in this report is from their answers to the survey. In general, the students were satisfied with the course, and they considered the invited lecturer a great choice. The students also said that the course was well organized, though they would prefer to know way more in advance when the lectures would be and they would also prefer to have it earlier in the semester. All the scheduling was dependent on the lecturer’s availability, which make it difficult to share with the students before. We will try to get the commitment from the external lecturers more time in advance.

NTNU:TIA:REST

Institutt for industriell økonomi og teknologiledelse

Department of Industrial Economics and Technology Management

Engelsk

Anbefalte forkunnskaper

This course is designed for PhD candidates within the fields operations research, finance and economics. Knowledge of machine learning methods is recommended. Such knowledge can be obtained through the course Introduction to machine learning and AI methods with economic applications. Programming skills are also needed.

Forkunnskapskrav

Admission to a PhD programme within operations research, or completed masters courses in optimization.

Faglig innhold

Advanced course in economic applications of machine learning and AI is an intensive PhD course offered through the project "COMPutational economics and optimization - Agents, Machines and Artificial intelligence" (COMPAMA). COMPAMA is developing an emerging interdisciplinary area in the borderland between economics, optimization, psychology, machine learning and AI with the main purpose to understand the economic impact of decisions, made by both machines and human agents.This course will extend the knowledge in machine learning methods applied to economics, going beyond the traditional unsupervised and supervised methods. The main goal is that the students can understand and apply sophisticated models in economic applications. Examples of relevant applications are algorithm trading, portfolio optimization and dynamic pricing.

Kursmateriell

Selected literature. Will be given at course start-up.

Læringsutbytte

After having completed the course the candidate should be able to:<ul><li>explain and implement the techniques learned;</li><li>choose the more suitable approach for a specific economic application;</li><li>recognize the opportunities and challenges of using AI in each context.</li></ul>

Læringsformer og aktiviteter

Lectures. Participation in the seminars is expected, which includes attendance at all lectures, as well as contributions to the discussions. There will be compulsory activities in the course.

Deltagelse og obligatoriske aktiviteter

Avansert kurs i økonomiske anvendelser av maskinlæring og kunstig intelligens

Bedriftsøkonomi og optimering

Bedriftsøkonomi

Økonomi og ledelse - doktorgrad (ph.d.)

Web/Lists(guid'8b698a2d-6ad0-4f7d-b3be-219cc1979f02')/Items(17052)/FieldValuesAsText

Avansert kurs i økonomiske anvendelser av maskinlæring og kunstig intelligens - IØ8813