Natascha Kljun

Linköping University in Sweden is offering a fully funded PhD position in Spatio-Temporal Machine Learning, with a focus on developing novel methods for analyzing and predicting spatio-temporal processes. The research will be conducted at the Division of Statistics and Machine Learning (STIMA), Department of Computer and Information Science, in collaboration with the Centre for Environmental and Climate Science at Lund University. The project is part of a collaborative initiative (ELLIIT) and will involve close cooperation between the two universities, including regular meetings and research visits. The PhD project centers on the development of generative models for grid-based and particle-based spatio-temporal data, controlled generation methods for data assimilation, and graph-based multi-scale neural network models. While the methods will be broadly applicable, there is a particular emphasis on inferring gas dynamics in urban environments, which is crucial for air quality management, climate change mitigation, and emergency response. The research aims to address current limitations in modeling approaches, such as computational inefficiency and lack of reliable uncertainty quantification, by enabling efficient and accurate inference and prediction at high spatial and temporal resolutions. Applicants should have a strong background in machine learning, statistics, computer science, fluid mechanics, or a related field, with a Master's degree or equivalent qualifications (at least 240 credits, including 60 advanced). Excellent academic results, strong mathematical skills, and documented experience in implementing models and algorithms are required. Candidates should be highly motivated for fundamental research, possess strong collaborative and communication skills, and be proficient in English. The position is fully funded, with salary and benefits according to Linköping University policy. The employment is typically for four years full-time, with possible extension up to five years depending on teaching and departmental duties. The successful candidate will be expected to contribute to the collaborative research environment and may have teaching or other departmental responsibilities up to 20% of full-time. Supervision will be provided by Prof. Fredrik Lindsten (Linköping University) and Prof. Natascha Kljun (Lund University). The application deadline is February 6, 2026. For more information and to apply, visit the official vacancy page. Key research areas: spatio-temporal machine learning, statistics, computer science, fluid mechanics, gas dynamics, AI, graph neural networks, data assimilation, urban air quality, and climate science.

Linköping University, one of Sweden's leading AI institutions, invites applications for a PhD position focused on the development of novel spatio-temporal machine learning methods. The research will address fundamental challenges in modeling and inferring gas dynamics in urban environments, with broad applications in air quality management, climate change mitigation, and emergency response. The project centers on advancing generative models for grid-based and particle-based spatio-temporal data, controlled generation methods for data assimilation, and graph-based multi-scale neural network models. Particular emphasis is placed on overcoming current limitations in modeling gas dynamics, such as computational inefficiency, inability to handle diverse sensory data, and lack of reliable uncertainty quantification. The successful candidate will join the Division of Statistics and Machine Learning (STIMA) within the Department of Computer and Information Science at Linköping University, collaborating closely with the Centre for Environmental and Climate Science at Lund University through the ELLIIT initiative. The project offers access to state-of-the-art computing infrastructure and a vibrant, collaborative research environment. Regular meetings and research visits between the two institutions will foster tight collaboration, with the advertised position focusing on method development and a parallel position at Lund University addressing applied aspects. Applicants should hold a Master’s degree in machine learning, statistics, computer science, fluid mechanics, or a related field, or have completed at least 240 credits with a minimum of 60 advanced credits in relevant subjects. Equivalent experience is also considered. Strong mathematical skills, documented experience in implementing models and algorithms, a drive for fundamental research, collaborative abilities, and excellent communication skills in English are required. The PhD position is full-time for four years, with possible extension up to five years based on teaching and institutional assignments. Salary is determined according to a locally negotiated progression, and employment benefits are provided. The starting date is by agreement. Applications must be submitted online by February 6, 2026, including a personal letter detailing suitability for the research topic. Linköping University values diversity and equal opportunities in its academic community. For further details and to apply, visit the official application page. We look forward to receiving your application and welcoming you to our innovative research environment.