AI Master's Thesis Awards 2021
AI Master's Thesis Awards 2021
Artificial Intelligence is a popular topic among students at NTNU. In recent years, around 200 master’s theses in which AI plays a substantial part have been submitted yearly. On December 3rd, we celebrated the top AI theses of 2021 in an awards ceremony at Gløshaugen. Sindre Stenen Blakseth, a 2021 graduate from the Department of Physics, was the well-deserved winner of the AI Master’s thesis awards, with his thesis: "Introducing CoSTA: A Deep Neural Network Enabled Approach to Improving Physics-Based Numerical Simulations".

A tough choice for the evaluation committee
In order to identify the best AI master’s theses in 2021, we reached out to the research community at NTNU to nominate excellent theses within AI. This resulted in the list of nominations that you can see to the right. The evaluation committee was made up by NAIL researchers and representatives from partners DNB and Telenor. Their job was to select the top three out of the 13 nominees, which was not an easy task, as the overall quality was very high.
The nominations covered a variety of topics, both within applied and theoretical AI. This year’s nominees were also a more diverse group of candidates than earlier. For instance, there were many more female students nominated to the awards than previous years. We’re happy to see this, as AI is still a male-dominated field. In addition, the nominees had different academic backgrounds and represented five different departments. In the end, the committee managed to boil it down to these three:
- Anna Haugsbø Hermansen - Machine Learning for Spatio-Temporal Forecasting of Ambulance Demand: a Norwegian Case Study (Dept of Computer Science)
- Sindre Stenen Blakseth - Introducing CoSTA: A Deep Neural Network Enabled Approach to Improving Physics-Based Numerical Simulations (Department of Physics)
- Anna Rodum Bjøru - The importance of disentanglement when learning representations (Dept of Computer Science)
The awards ceremony
Anna, Sindre and Anna were invited to present their work during the awards event that took place at Gløshaugen. Before the nominees presented their work, Massimiliano Ruocco, research scientist at SINTEF and adjunct associate professor at NTNU, gave a talk on his experience working with companies and academia on industrial master’s theses. We also heard from the two student organisations, BRAIN and Cogito, who presented their activities and plans for 2022.
All nominations
- Ida Merete Enholm - Responsible AI Governance and Its Effect on Competitive Performance
- Jon Riege and Klara Schlüter - Stochastic Multiplicative Updates for Nonnegative Matrix Factorization
- Anna Rodum Bjøru - The importance of disentanglement when learning representations
- Halvor Ødegård Teigen, Vebjørn Malmin - Reinforcement Learning and Predictive Safety Filtering for Floating Offshore Wind Turbine Control – A step towards safe AI
- Nicolò Oreste Pinciroli Vago - Identification of salient iconography features in artwork analysis
- Mathias Backsæther - Silent Speech Communication Using Facial Electromyography
- Anna Haugsbø Hermansen - Machine Learning for Spatio-Temporal Forecasting of Ambulance Demand: a Norwegian Case Study
- Anja Rosvold From & Ingvild Unander Netland - Fake News Detection by Weakly Supervised Learning: A content-based approach
- Sindre Stenen Blakseth - Introducing CoSTA: A Deep Neural Network Enabled Approach to Improving Physics-Based Numerical Simulations
- Vemund Fredriksen & Svein Ole Matheson Sevle - Pulmonary Tumor Segmentation Utilizing Mixed-Supervision in a Teacher-Student Framework
- Fredrik Foss and Truls Stensrud - Foraging, Genetic Network Programming and its Hybridization with NEAT: Evolving Evolutionary Swarm Robotics
- Simen Burud - Conversational Language Models for Low-Resource Speech Recognition
- Silas Eichsteller - Multi-image detection and tracking of cracks in ship tanks
Heri Ramampiaro, professor in AI and Head of the Computer Science Department, reflected upon the role of NAIL in bringing students, researchers, business and industry together. The AI Master’s Thesis Awards is one example of an initiative that promotes this. Moreover, the awards is a concrete result of the collaboration between NAIL and our partners, specifically Telenor which is the sponsor of the event. This is just one example of Telenor's commitment to AI students at NTNU.
And the winner was..
After excellent presentations from all three, it was announced that Sindre Stenen Blakseth made it to the top. Sindre's thesis introduces a new Hybrid Analysis and Modelling (HAM) approach denoted the Corrective Source Term Approach (CoSTA) where a Deep Neural Network is used to compensate for the deficiencies of a pure physics-based approach. The work may enable safe AI with a built-in sanity check mechanism. Furthermore, in the context of predictive digital twin CoSTA enables the model to evolve with the asset during its life cycle. Sindre received a prize of NOK 20 000, while Anna and Anna received 10 000 NOK prizes.
Congratulations to the winners and all nominees!