Postdoctoral Position in Machine Learning for Odor Perception KTH Royal Institute of Technology in Sweden

This postdoctoral position at KTH Royal Institute of Technology offers an exciting opportunity to join the EU project “Digitising Smell: From Natural Statistics of Olfactory Perceptual Space to Digital Transmission of Odors.” The project aims to digitalize the sense of smell, advancing our understanding of olfaction in humans and enabling the development of AI models that simulate olfactory experiences. As a postdoctoral researcher, you will focus on processing and developing representation models for diverse data sources, including time-series data (EEG, video, mass spectrometry), text, images, and chemical data (molecular graphs, SMILES strings) related to odorant stimuli.

The research environment is highly interdisciplinary, bridging machine learning, neuroscience, and chemistry. You will work with deep learning architectures such as Transformers, diffusion models, and graph neural networks, applying them to multimodal and high-dimensional data. The project emphasizes multimodal representation learning and integration of heterogeneous data sources, requiring expertise in signal processing, statistical modeling, and advanced programming skills in Python and ML frameworks (PyTorch, TensorFlow, JAX).

KTH Royal Institute of Technology is a leading international technical university located in Stockholm, Sweden. The institution is committed to education, research, and innovation, with a strong focus on sustainability, equality, diversity, and equal opportunities. The postdoctoral appointment is full-time and temporary, lasting up to two years, and offers a monthly salary with attractive benefits and good working conditions.

Applicants must hold a doctoral degree or an equivalent foreign degree, obtained within the last three years prior to the application deadline. Required qualifications include proven research experience with deep learning architectures applied to multimodal data, expertise in time-series modeling and chemical/structural data representation, familiarity with multimodal representation learning, signal processing, and statistical modeling of high-dimensional data, and strong programming skills. The ability to work collaboratively in interdisciplinary teams and excellent communication skills are essential. Teaching and supervision experience is an advantage. Applicants must submit diplomas and grades, with translations into English or Swedish if necessary.

To apply, log into KTH's recruitment system and submit your application online. Include your CV, diplomas and grades (with translations if needed), and a brief account of your research motivation and academic interests. Ensure your complete application is received by midnight CET on the deadline date, 2026-06-27. For further information, contact Professor Danica Kragic Jensfelt at [email protected].

This position is a unique opportunity to contribute to cutting-edge research in olfactory modeling and machine learning, within a creative and dynamic academic environment at KTH.