PhD student position at Umeå University in the NEST project STING (Synthesis and Analysis with Transducers and Invertible Neural Generators).
Project description
The focus of the project is graph based machine learning for discrete and continuous structures, led by Frank Drewes, Johanna Björklund, and Henrik Björklund.
The PhD student will work within the context of the STING (Synthesis and Analysis with Transducers and Invertible Neural Generators) project, financed by WASP (Wallenberg AI, Autonomous Systems and Software Program). The project conducted in collaboration with the Royal Institute of Technology (KTH) in Stockholm and Linköping University.
Human communication is multimodal in nature, and occurs through combinations of speech, language, gesture, facial expression, and similar signals. STING aims to design models that capture this richness, uniting synthesis and analysis with the help of both discrete and neural models. Covering both synthesis and analysis in one framework also allows us to create efficient mechanisms for explainability, and to inspect and enforce fairness in the models.
We ourselves are experts on discrete models for graph generation, and have recently started to apply these to the task of parsing multimodal language data. We also work on bias analysis and mitigation. The partner research groups bring complementary expertise to the project: KTH has extensive experience with probabilistic deep learning for analysis and synthesis of human communication. Linköping University complements these aspects with in-depth knowledge of natural language processing.
In addition to its scientific value, the technologies developed by the project are expected to have a substantial societal imprint. Read more about STING
The research focus of the doctoral student within the spectrum covered by STING will be determined based on the scientific background and interests of the selected candidate. The project is conducted within the research group for the Foundations of Language Processing at Umeå University. The group studies theoretical and practical aspects of representing language on computers, and its interconnection with other sources of information. The work of the group spans from formal language theory to applied natural language processing. The group consists of five senior researchers and eight PhD students.
More information is available at: https://www.umu.se/en/research/groups/foundations-of-language-processing/