Doctoral Student in Urban Heat Prediction Using AI and Earth Observation

Project description

Third-cycle subject: Computer Science

The Division of Robotics, Perception and Learning has an open position for a doctoral student with a background and strong interest in deep generative learning and computer vision/remote sensing. The successful candidate will join a project funded by Digital Futures to map urban air temperature at high spatial and temporal resolutions. Cities face rising urban heat, both from sharp spikes and prolonged hot periods; however, producing these high-resolution urban air temperature maps remains an open problem. This project will treat dense urban air temperature estimation as an inverse problem, using conditional diffusion models to fuse satellite, forecast, and crowd-sourced station data into probabilistic maps. The challenge: conditioning generative models on large, heterogeneous, incomplete data will require the student to work on cutting-edge ML for a pressing societal need.

You will be part of an ongoing collaboration between the groups of Assoc. Prof. Josephine Sullivan at RPL, Prof. Yifang Ban (Division of Geoinformatics) at the Department of Urban Planning and Environment, and Sebastian Hafner at RISE, Kista.

Supervision: Associate Professor Josephine Sullivan is proposed to supervise the doctoral student.

Admission requirements

To be admitted to postgraduate education (Chapter 7, 39 § Swedish Higher Education Ordinance), the applicant must have basic eligibility in accordance with either of the following:

• passed a second cycle degree (for example a master's degree), or

• completed course requirements of at least 240 higher education credits, of which at least 60 second-cycle higher education credits, or

• acquired, in some other way within or outside the country, substantially equivalent knowledge.

• This project will require practical proficiency in deep learning. Thus demonstrated competency in deep learning programming libraries such as TensorFlow, PyTorch, or JAX is a must and experience with GPU-based experimentation and cluster computing (e.g., Docker, Slurm) a plus.

In addition to the above, there is also a mandatory requirement for English equivalent to English B/6.

Selection

In order to succeed as a doctoral student at KTH you need to be goal oriented and persevering in your work. During the selection process, candidates will be assessed upon their ability to:

• independently pursue his or her work

• collaborate with others,

• have a professional approach and

• analyze and work with complex issues.

In the evaluation of candidates, an emphasis will be placed on academic results, completed courses and demonstrated programming ability via completed project work. An earlier specialization in machine learning and experience with computer vision and/or remote sensing is highly desirable and especially meritorious.

After the qualification requirements, great emphasis will be placed on personal skills.

Employment

Full-time doctoral student employment in Stockholm (start according to agreement). Total employment may not exceed four years of full-time doctoral education. Monthly salary according to KTH's doctoral student salary agreement.

To apply

Apply through KTH's recruitment system. Applications must include:

• Copies of diplomas and grades from previous university studies and certificates of fulfilled language requirements. Translations into English or Swedish if the original is not issued in one of these languages. Copies of originals must be certified.

• CV including relevant professional experience and knowledge.

• Application letter (max 2 pages) describing why you want to pursue research studies, your academic interests, and how they relate to your previous studies and future goals.

• Representative publications or technical reports. For longer documents, provide a summary (abstract) and a web link to the full text.

Last application date: 24 September 2026, midnight CEST. Reference: PA-2026-2917.