The project Waste Values concerns​​ transformations of waste in the circular economy. We live in a time when waste is expected to be given new life as it gets upgraded and enters new cycles. But value transformations require work and the circular economy is far from automated. In addition to functioning infrastructures, organizations, economic management and policy, changed knowledge, action and meaning-making are required for something that was once considered worthless to be revalued. The project studies the processes through which value transformations of bio-waste are made. Examples of wastes that might be of interest are wastewater, stormwater or forest waste. The project is led by Professor Tora Holmberg.

The person who receives the position will be supervised by the project manager but work independently and has great opportunity to influence and design the project, collect and analyze relevant data.

We are an interdisciplinary team of experimentalists, analysts and modelers, who collaborate to address important questions of malaria parasite development and transmission by mosquitoes. We are now looking for an experimental geneticist or an cryo-imaging expert for projects funded by the Wallenberg Foundation. Aim of the position is to study how molecular interactions between malaria parasite and the mosquitoes that transmit them determine the reproductive rate of malaria. In search of new transmission blocking strategies, we will work in an international collaboration to combine genome-scale genetic screens, single cell transcriptomics, metabolomics and metabolic modelling to gain a global mechanistic understanding of malaria transmission and determinants of vector competence. The lab has a strong track record of performing insightful genetic screens in different life cycle stages (e.g. Bushell et al., 2017; Stanway et al., 2019; Russell et al., 2023). The primary aim of the current project is to evolve and apply these powerful technologies for a deeper understanding of parasite-mosquito interactions.

Our research team explores the molecular underpinnings of myocarditis through a systems medicine approach. By integrating data from genomics, virology, immunology, metabolomics, and microRNA analyses, all derived from well-characterized patient cohorts, we aim to uncover molecular signatures linked to disease onset, severity, and the risk of long-term cardiac complications such as dilated cardiomyopathy. 

The project will develop privacy-aware machine learning (ML) models. We are interested in data-driven models for complex data, including high-dimensional data. We are interested in both data-driven models as well as models built from synthetic data. Within privacy, we are interested in different types of privacy measures and models (differential and integral privacy, k-anonymity), different scenarios (centralized and decentralized data; local and global privacy). For decentralized data, we consider federated learning. We are particularly interested in transparent models.

The Privacy-aware transparency decisions research group (led by Prof. Vicenç Torra) conducts research in data privacy for data to be used for machine and statistical learning. It is well known that data can be highly sensitive, and that naive anonymization is not sufficient to avoid disclosure. Models and aggregates can also lead to disclosure as they can contain traces of the data used in their computation. We want to understand the fundamental principles that permit us to build privacy-aware AI systems, and develop algorithms for this purpose. The group collaborates with several national and international research groups, edits one of the major journals on data privacy (Transactions on Data Privacy), and has active links with the private and public sectors.

Linear algebra expressions are evaluated in an efficient and robust way by mapping them to a carefully chosen sequence of calls to optimized functions as offered by libraries such as BLAS and LAPACK. The mapping is by no means unique, and different mappings differ in terms of time, space, and accuracy. When the sizes of the matrices are unknown at compile-time, as is often the case, then the problem is further complicated by the fact that no single mapping is optimal for all combinations of matrix sizes. As a consequence, any code generated (at compile-time) to evaluate the at run-time) must consist of more than just a single sequence of function calls.

We recently received a grant from the Swedish Research Council to investigate how to effectively compile linear algebra expressions when the matrix sizes are unknown at compile-time. The project aims to address the problem using e-graphs. An e-graph is a data structure commonly used in automated theorem provers and recently popularized for optimizing compilers and many other optimization tasks. We hypothesize that an e-graph would be an effective tool for discovering and encoding a large set of alternative ways of evaluating a particular expression with unspecified matrix sizes. When a concrete expression is evaluated at run-time, thus revealing the matrix sizes, an extraction algorithm can identify an optimal evaluation scheme for that particular instance of the expression from the compiled e-graph.