Projects

The IDEA Project focuses on digitalisation and predictive justice, which can be applied in labour law disputes, both in European and national contexts. We aim to develop a digitalised procedure and a platform to address harmonisation across digitalisation and predictive algorithms in access to justice, encompassing interaction with the forthcoming EU legislations on the Digitalisation of judicial operation and access to justice and the AI Act and to verify it together with six judicial authorities and numerous practitioners. This method has an innovative character – combining legal analysis with broad text mining techniques – as it will unravel the concept of predictivity to assess the (in)compatibility with the rule of law in civil law systems.

CREA3 aims to introduce Artificial Intelligence (AI) driven tools to assist EU citizens in resolving disputes by applying innovative game theoretical (GT) algorithms. To improve access to justice and interaction with the rule of law, we seek to develop a strategy that addresses harmonisation across digitalisation and predictive algorithms. This approach, which combines legal research with extensive text-mining techniques, is novel since it will explore the idea of testing and evaluating how well family law systems in six EU legislations adhere to the new forthcoming “Regulation on the digitalisation of judicial cooperation and access to justice in cross-border civil, commercial and criminal matters”.

How do neurons maintain their function throughout life? Neurons are born early in development and are not regenerated during an individual’s lifetime. Therefore, after maturation, our neurons must remain functional throughout life, which in humans can be up to a hundred years of more. The goal of my ERC Starting Grant 2024 proposal EpiNeuroLife was to uncover how the formation of the epigenetic landscape during neuronal maturation contributes to the maintenance of neuronal gene expression and function during adulthood and ageing. My preliminary studies show that developing neurons in the mouse cerebellum accumulate extremely high levels of the repressive histone modification H3K27me3 during maturation, which I hypothesise is critical for regulating chromatin compaction and neuronal gene expression later in life. In MOB3ERC113, I will generate supporting data for my ERC application to uncover the mechanisms of H3K27me3 deposition in neurons.

The overall objective of the EquiTech project is to increase the capacity of Estonian and Lithuanian authorities to address risks of bias and discrimination in ADM procedures. This will be achieved through the development of research, support materials, training and a media campaign. This will be achieved as follows: 1. A desk study addressing gaps in the Estonian and Lithuanian policy and legal frameworks on AI, together with a mapping of potential risks of discrimination and bias in ADM systems used by Estonian authorities. 2. Developing practical tools to prevent potential risks of discrimination and bias in the design and implementation of ADM systems; 3. Improving the skills and knowledge of users and developers of ADM systems through training, study tours, an international conference and academic webinars; 4. Raising public awareness of the risks of discrimination and bias in public sector ADM systems. As a result, we will reduce the potential for discrimination and prejudice in the Estonian and Lithuanian ADM systems and set an example for other Member States by contributing to national anti-racism action plans and strategies against anti-Semitism, anti-Muslim hatred and xenophobia, LGBTIQphobia and all other forms of intolerance.

Advancing education of engineers through AI for cyber-sustainable and energy-efficient union involves a collaborative ENERGYCOM network of Baltic, Scandinavic and Nordic institutions, including Kaunas University of Technology, Vilnius Gediminas Technical University, Riga Technical University, Tallinn University of Technology, University of Tartu, University of Vaasa and UiT The Arctic University of Norway. This network aims to address engineering education, focusing on cyber-sustainability and energy efficiency by integrating advanced AI-powered teaching and learning methods. Within this project, ENERGYCOM introduces AI-powered assistants into the learning ecosystem, preparing engineers to tackle global challenges in sustainability and cybersecurity while adhering to ethical principles. Key activities include workshops, express mobility programs, and academic exchanges covering fields such as like mechatronics, power systems, robotics, cybersecurity, artificial intelligence, and human-centric information systems. By integrating AI-driven tools into education, the initiative aims to prepare students and educators to think critically, solve complex problems, and develop sustainable and energy-efficient solutions. The network builds on the EU's strategic initiatives, such as the Digital Education Action Plan, Coordinated Plan on Artificial Intelligence and Cybersecurity Strategy, to strengthen digital competencies and promote sustainability.

Although there is a clear need in Estonia to make economic policy more responsible and smarter, the call for projects highlights the challenge of a lack of unified understanding of what sustainability entails across different sectors in the long term. Most of the described pain points are also noted in strategic documents guiding Estonia’s sustainable economic development. Hence, the key problems are more evident at the policy implementation level, where public sector interventions have sometimes yielded opposite results, had minimal impact, or progressed too slowly. The project aims to analyze and solve these problems through the following activities First, to analyze and synthesize what opportunities does the green transition offer for Estonia, and what prerequisites are needed to realize them? We will validate some of these opportunities, create cross-sectoral links, and obtain a realistic picture of what the green transition entails in sectors most crucial for sustainability transition. Second, clarifying the above and mapping the way forward will allow for a deeper investigation into necessary changes in policy measures: which public sector interventions should be increased, newly created, reduced, or discontinued. Third, to achieve real change in businesses, we will collaborate with the Client to design novel interventions for the next period's economic policy. We expect long-term results, as the project coincides with policy design for the 2027-2034 budget period, enabling the public sector to acquire skills that are more effective for solving today’s and future problems. The output will be policy recommendations (measure design and skills) implemented from 2027 onward.

This project studies new dependently typed systems suitable for the development and mechanization of programming language semantics. Particular emphasis is given to languages supporting concurrency and non-determinism, such as transition systems and process calculi. Popular proof assistants based on dependent type theory, such as Agda and Coq, are inadequate for the formal verification of the denotational semantics of such languages because of their insufficiently expressive type systems. We address this issue by extending modern type-theoretic frameworks, such as homotopy type theory, with a new class of coinductive types coming from final coalgebras of accessible functors. In denotational semantics, these types are necessary for handling the non-deterministic and continuously-interactive behavior of processes. The resulting more expressive type systems will prove themselves capable of encoding the formal semantics of various languages with concurrency and non-determinism.

A platform for monitoring energy usage and operational state properties of distribution networks is elaborated, enabling users of distribution networks to access relevant and better time-resolution data to support green energy usage, energy efficiency and the flexibility technologies. The platform being developed will provide extensive support for decision-making and calculation algorithms, enabling efficient availability and management of local area operational data.

A cloud-based digital health platform for private clinics in East Africa will be developed, focusing on optimizing workflows, monitoring chronic patients, and managing patient care. A distinctive feature of the platform is the combination of a mobile application and an SMS-based triage solution, allowing access for patients with limited digital capabilities. The product is being piloted in Kenya, where a shortage of doctors and a growing population create strong demand for efficient, labor-saving solutions. The solution is based on the logic of primary care digital models tested in Estonia, adapted to the local context.

The project will develop a smart, adaptive electric drive that increases the energy efficiency and reliability of electric vehicles. The innovative solution combines artificial intelligence-based control with advanced sensor technology, allowing the drive to adapt in real-time to changes in traffic and road conditions. The project will produce a laboratory prototype, a user-friendly software solution for data processing, and comprehensive documentation that will simplify the implementation of the system.

Fish-AI is a real-time automatic video-based fish monitoring system developed by TalTech that automates reporting in accordance with EU requirements and significantly reduces labor costs. The system is compatible with various underwater cameras, detects different water conditions, and is able to adapt to them.

The project will develop a computer simulation model that will enable the assessment of fish passage conditions through hydroelectric power plant turbines. The use of simulations will significantly reduce the need for field tests, allowing for more efficient use of time and money and a broader analysis of scenarios.