dc.date.accessioned | 2023-06-01T09:40:11Z | |
dc.date.available | 2023-06-01T09:40:11Z | |
dc.date.issued | 2023 | |
dc.identifier.uri | http://hdl.handle.net/10852/102387 | |
dc.description.abstract | Artificial intelligence and machine learning are increasingly used in many industries and applications, where more and more important decisions are being made based on machine learning models. However, recent research has shown limitations to many of the commonly used machine learning methods, such as lack of stability and transparency. As machine learning models are increasing their impact on many areas of people's lives, it is critical to ensure safe and responsible use of these methods.
This thesis addresses challenges related to the use of machine learning for safety-critical systems. These systems often involve complex physical relationships with a large degree of uncertainty and time dependency. An example of such a system is the Earth’s weather dynamics, where we explore it in the context of safe airplane landings and extreme weather phenomena.
The exploration is carried out by considering different ways to create more trustworthy machine learning models. This includes training the models to follow physical laws and mathematical constrictions, so that they behave according to the rules of system. We also explore how the models can provide explanations of their own behavior and information about their uncertainty, so that we know how they work and when we can trust them. | en_US |
dc.language.iso | en | en_US |
dc.relation.haspart | Paper I. Midtfjord, A. D. and Huseby, A. B. “Estimating Runway Friction Using Flight Data”. In: Proceedings of the 30th European Safety and Reliability Conference and the 15th Probabilistic Safety Assessment and Management Conference. (2020), doi: 10.3850/978-981-14-8593-0. The article is included in the thesis. Also available at: https://doi.org/10.3850/978-981-14-8593-0 | |
dc.relation.haspart | Paper II. Midtfjord, A. D., De Bin, R. and Huseby, A. B. “A decision support system for safer airplane landings: Predicting runway conditions using XGBoost and explainable AI”. In: Cold Regions Science and Technology. Vol. 199, no. 103556 (2022), doi: 10.1016/j.coldregions.2022.103556. The article is included in the thesis. Also available at: https://doi.org/10.1016/j.coldregions.2022.103556 | |
dc.relation.haspart | Paper III. Midtfjord, A. D., De Bin, R. and Huseby, A. B. “A copula-based boosting approach for time-to-event prediction”. Submitted for publication. To be published. The paper is not available in DUO awaiting publishing. | |
dc.relation.haspart | Paper IV. Midtfjord, A. D., Eggen, M. D. “Delay-SDE-net: A deep learning approach for time series modelling with memory and uncertainty estimates”. Submitted for publication. To be published. The paper is not available in DUO awaiting publishing. | |
dc.relation.uri | https://doi.org/10.3850/978-981-14-8593-0 | |
dc.relation.uri | https://doi.org/10.1016/j.coldregions.2022.103556 | |
dc.title | Machine learning methods for safety-critical systems with time dependency: With applications to airplane landings and environmental data | en_US |
dc.type | Doctoral thesis | en_US |
dc.creator.author | Midtfjord, Alise Danielle | |
dc.type.document | Doktoravhandling | en_US |