Postdoc on Causal Machine Learning for Spatio-temporal Datasets

at  Universiteit Leiden

Vacaturenr.
15086
Functie-categorie
Wetenschappelijke functies
Omvang (fte)
1,0
Extern/intern
Extern
Locatie
Leiden
Geplaatst op
13 augustus 2024
Sluitingsdatum
30 september 2024
The Faculty of Science and the Leiden Institute of Advanced Computer Science (LIACS) are looking for a:

POSTDOC ON CAUSAL MACHINE LEARNING FOR SPATIO-TEMPORAL DATASETS

The Leiden Institute of Advanced Computer Science (LIACS) is looking for an excellent Postdoc researcher (1.0 FTE) with a background in Computer Science (or a closely related field) to join a project focused on developing an advanced machine learning framework for spatio-temporal datasets. The position is for 4 years and is partially funded by the Dutch Research Council (NWO) through the Aspasia premium awarded to dr. Mitra Baratchi for her research on machine learning for spatio-temporal data. The successful applicant will be embedded in the Spatio-temporal data Analysis and Reasoning (STAR) and the Automated Design of Algorithms (ADA) research groups and collaborate with researchers at the Natural Computing (NACO) and the Explanatory Data Analysis (EDA) Research Groups. Moreover, there are plenty of opportunities for interaction and collaboration with other groups at the institute and internationally.
The research project is broadly focused on automated intervention design. An intervention is any external interference in an ongoing process that is performed to achieve a particular outcome. Being able to answer interventional questions automatically will help decision-makers solve complex challenges (i.e., reducing methane emissions or reducing the infection rate during a pandemic). Most machine learning algorithms designed for spatio-temporal data do not allow assessing the impact of interventions without capturing causal links. At the same time, configuring causal machine learning algorithms is extremely difficult, being an unsupervised learning task. In this project, we aim to design algorithmic solutions (e.g., new Automated Machine Learning approaches) for automatic and scalable assessment of interventions based on observational spatio-temporal datasets generated by modern sensing technologies (e.g., IoT sensors, mobile devices, Earth observations).

SELECTION CRITERIA

We are looking for a candidate with expertise or experience with one or more of the following topics: machine learning for spatio-temporal data, causal machine learning (casual discovery and causal inference), and automated machine learning.

• holding (or close to acquiring) a PhD degree in Computer Science, Artificial Intelligence or a closely related field;
• strong research vision and an academic mindset;
• strong publication record;
• being able to collaborate with scientific peers inside and outside your own research area;
• strong programming skills in Python or other programming languages;
• excellent proficiency and communication skills in English;

Within this position, it is expected that you will:

• conduct research within the scope of the project leading to peer-reviewed publications in journals and conference proceedings;
• collaborate closely with dr. Elena Raponi (the Natural Computing Group) and dr. Saber Salehkaleybar (the Explanatory Data Analysis Group) at LIACS;
• engage in teaching activities of the STAR Research Group and supervision of BSc and MSc students;