PhD position: research on an AI Buddy for instructional feedback

(07-10-2022)

We live in a society that is driven by computer science, artificial intelligence, data science and STEM. It is time to adapt our educational system to shape ‘creators of the future’! Together with you we want to make programming and computational thinking more accessible in the classroom. More specifically, with AI, we want to give teachers insight in the learning process of their students.

However, observing the learning process is difficult and costly, especially in an open-ended STEM context. Moreover, it is often challenging for teachers to effectively assess the learning process. The goal of this PhD research project is to develop a tool that can monitor the progress of students in an open-ended learning context.

A unique interdisciplinary research opportunity

This research project is an interdisciplinary project that involves two labs: IDLab at UGent-imec and Itec KU Leuven-imec. Your PhD project will be led by two PI’s, prof. Francis wyffels (IDLab-AIRO) and prof. Fien Depaepe (Itec), who each bring their own expertise to this project. Furthermore, you will be able to collaborate with other enthousastic researchers of both labs!

IDLab is a research group of UGent, as well as a core research group of imec. IDLab performs fundamental and applied research on data science and internet technology, and counts over 300 researchers. Our major research areas are machine learning and data mining; semantic intelligence; multimedia processing; distributed intelligence for IoT; cloud and big data infrastructures; wireless and fixed networking; electromagnetics, RF and high-speed circuits and systems. The AIRO (“AI and robotics” - https://airo.ugent.be) team of IDLab has been studying various kinds of neural networks for more than 20 years. Our focus is on applications in robotics and in human-AI interaction. AIRO has been at the forefront of deep learning research ever since it became popular a decade ago. Illustrative of this success are an excellent track record at Kaggle competitions. thanks to a consistent focus on innovative ideas that are somewhat outside the mainstream “AI-hype”, many of our former PhD students are now working at top commercial research labs, such as Deepmind or Google Brain, or hold strategic senior R&D positions in companies.

Itec is a research group of KU Leuven and imec, Flanders' high-tech research and innovation centre for nanoelectronics and digital technologies. At KU Leuven's Kortrijk campus, itec brings together researchers from three different faculties (Psychology & Educational Sciences, Arts and Medicine) and five disciplines (educational psychology, statistics, applied linguistics, data science and computer science) to collaborate on research topics in educational technology, such as instructional design and the effectiveness of online learning environments (e.g. e-learning, serious games), multilevel modelling, algorithms for adaptive learning and natural language processing. Itec is structurally part of imec and takes charge of the research program imec Smart Education, within which a broader inter-university team of research groups interacts (also from UGent and VUB). The imec Smart Education research program focuses on research on the use of technologies such as learning analytics, artificial intelligence or wearables. 

Your tasks

You will research, develop and validate AI techniques for monitoring the progress of students in an open-ended learning context. More specifiically, you will focus on code representation, self-supervised visualization techniques, data visualization for a dashboard with feedback on the students’ progress, and validation of the dashboard with teachers. Due to the collaboration with Dwengo vzw, you will be able to see everything integrated in their open-source online programming and simulation environment, https://blockly.dwengo.org, which is used by more than 1000 students every month. Consequently, large-scale validation will also be possible.

Code representation

  • Take advantage of the logging system currently implemented in the programming environment (blockly.dwengo.org – currently 1000 users/month) to collect additional data about the programming process of students.
  • Distribute teaching materials to encourage teachers to use our programming environment.
  • Develop proof of concept teacher dashboard with visualizations of the solution paths.

Self-supervised visualization techniques

  • Use self-supervised machine learning techniques to visualize learner's solution paths to teachers.
  • Predict the next step in a solution path of a learner and use it to detect paths diverging from a correct solution.
  • Integration of the developed techniques into a teacher dashboard in collaboration with Dwengo vzw.

Dashboard with feedback on students’ learning process

  • Understand which visualizations are useful for teachers to better support learning processes.
  • Evaluate how teachers perceive, interpret and make decisions based on the data presented on the dashboard.

Validation

  • Evaluate teachers' beliefs towards using the teacher dashboard.
  • Observe how teachers use the teacher dashboard in actual classroom practices.

Who we are looking for

The ideal candidate should be a talented programmer with expertise in machine learning and an interest in educational science. We value candidates with:

  • An academic Master’s degree or equivalent, in any of the following branches: Computer science, Applied Engineering in computer science, Computer Science Engineering or equivalent.
  • Excellent knowledge and experience in programming in Python.
  • Experience with both front- and backend JavaScript is desirable.
  • Hands-on experience with machine learning is a big plus.
  • Experience with education or teaching experience is a big plus.

Candidates are expected to have good communication skills and proficiency in written and oral English. Since the you will interact with Flemish teachers, B2 (or higher) profinciency in Dutch/Flemish is required.

In addition, a problem-solving attitude and the ability to take initiative and work independently to deliver high-quality research is essential for a successful completion of the project. Since this is an interdisciplinary project, we also expect the candidates to be capable of critical self-reflection, demonstrate an eagerness to learn, and be a team-player.

 What we can offer you

  • Full-time PhD scholarship for 4 years (coupled to a positive evaluation after the first year) at Ghent University.
  • A joint (UGent and KU Leuven) doctoral degree upon completion.
  • Support by experienced researchers and educators.
  • Support of Dwengo vzw, a non-profit organisation with whom we collaborate.
  • Your remuneration will be determined by salary scale WM1. Click here for more information about our salary scales.
  • All Ghent University staff members enjoy a number of benefits, such as a wide range of training and education opportunities, 35 days of holiday leave (on an annual basis for a full-time job) supplemented by annual fixed bridge days, a bicycle allowance and eco vouchers. Click here for a complete overview of all the staff benefits (in Dutch).

Interested?

This position is open and immediately available. We will review applicants on a rolling basis until we find the best candidate. However we set a final application deadline for the end of November, 2022. We will not accept applications after this deadline.

Your application must include the following documents:

  • your CV (in English or Dutch) and an overview of your study results
  • your application/motivation letter (in English or Dutch, max 2 pages)
  • a transcript of the required degree (if already in your possession). If you have a foreign diploma in a language other than our national languages (Dutch, French or German) or English, please add a translation in one of the mentioned languages.
  • an overview of your study results (bachelor’s and master’s)
  • You may also add other relevant documents such as a reference letter, certificate of employment, etc.

Please send your completed application (as a single pdf file, maximum size of 10MB) to Prof. Francis wyffels (Francis.wyffels@UGent.be). Please set the subject of your email as “AI Buddy – Application: <firstname> <lastname>”. We accept applications written in English or Dutch.

Ghent University maintains an equal opportunities and diversity policy, so we encourage applications from qualified individuals from different backgrounds.


More information

If you need more information about this vacancy, please contact Prof. Francis wyffels (Francis.wyffels@UGent.be) and Prof. Fien Depaepe (fien.depaepe@kuleuven.be). Please set your subject as “AI Buddy – Query”. You can address your questions in both, English or Dutch.