Doctoral Student in Urban Heat Prediction Using AI and Earth Observation
Develop cutting-edge machine learning and conditional diffusion models to map urban air temperatures using earth observation data.
Best for:Best for computer science graduates with a strong background in deep learning, generative models, and computer vision or remote sensing.
Overview
KTH Royal Institute of Technology invites applications for a doctoral student position focusing on computer science, deep generative learning, and remote sensing. Based in the Division of Robotics, Perception and Learning, the successful candidate will tackle urban heat challenges by treating air temperature estimation as an inverse problem using conditional diffusion models. The project integrates satellite observations, weather forecasts, and crowdsourced data to generate high-resolution probabilistic urban heat maps. The candidate will work within a collaborative environment involving KTH and RISE researchers, combining advanced machine learning with impactful societal applications.
What you'll do
- Design and implement conditional diffusion models to fuse satellite, forecast, and crowdsourced station data into probabilistic urban temperature maps.
- Handle large, heterogeneous, and incomplete environmental datasets using advanced machine learning techniques.
- Conduct GPU-based experimentation and utilize cluster computing infrastructure.
- Collaborate across research groups including the Division of Geoinformatics and RISE, Kista.
- Contribute to departmental tasks such as limited teaching or administration up to a maximum of 20% of working time.
- Author scientific publications and progress toward a doctoral degree in computer science.
Requirements
Degree required: Master's degree or equivalent second-cycle higher education qualification
Degree offered: PhD
Language: English equivalent to English B/6
Instruction: English
- Practical proficiency in deep learning
- Demonstrated competency in deep learning programming libraries such as PyTorch, TensorFlow, or JAX
- Strong background in computer science or related quantitative fields
- Ability to work independently and analyze complex technical problems
- Passed a second-cycle degree, completed at least 240 higher education credits with 60 second-cycle credits, or acquired equivalent knowledge
- Must be admitted to postgraduate education at KTH
Funding & terms
How to apply
Submit your application through the official KTH portal by following the online instructions provided on the position listing page. Ensure all required academic transcripts, project work examples, and supporting documents are attached before the closing date.
- CV
- Cover letter
- Master's degree certificate and transcripts
- Project work documentation demonstrating programming ability
Contact
Josephine Sullivan· Associate Professor
Create a free profile to unlock your personal match score, the full requirements, funding details and how to apply — in English or Persian.