Dr Ferreira has extensive experience in assessing the uncertainty of machine learning models and their integration with domain-specific knowledge, ultimately influencing critical decision-making processes. She is passionate about building confidence in model-based predictions though the combination of physics- and data-driven models.
About Dr Ferreira
Dr Ferreira specializes in assessing the uncertainty of machine learning models and their integration with domain-specific knowledge, ultimately influencing critical decision-making processes.
Her current focus is bridging cutting-edge research with real-world solutions. She is a firm believer in the power of collaboration, fostering inclusive teams where diverse perspectives drive innovation. Carla's dedication to diversity, inclusion, and equity in science stems from her conviction that it is essential for tackling multifaceted challenges. Her work has a significant impact on industries ability to develop and deploy AI responsibly, ethically and for the benefit of society. By combining data-driven and physical models, she maximizes the strengths of both approaches, enhancing confidence in the application of AI models.
Her expertise spans statistical approximation, uncertainty quantification and the financial impact of data-driven reservoir management.
I am deeply motivated by the potential of AI to solve real-world problems, but also acutely aware of the inherent uncertainties and risks. Working with AI/digital trust allows me to combine my passion for research and collaboration with my expertise in uncertainty quantification to ensure AI is developed and deployed responsibly, ethically and for the benefit of society.
Dr Carla Janaina Ferreira
Vessel Technical Index (VTI): conducted uncertainty quantification for the Vessel Technical Index (VTI), which evaluates a ship's technical performance considering relevant sources of uncertainty, enabling the users to take informed decisions based on the VTI calculations.
RaPiD - Reciprocal Physics-based and Data-driven models:This project aims to provide more specific, accurate and timely decision support in operation of safety-critical systems, by combining physics-based modelling with data-driven machine learning and probabilistic uncertainty assessment.
El Mekkaoui, S., Ferreira, C.J., Guevara Gómez, J.C., Agrell, C., Vaughan, N.J. and Heggen, H.O. (2023) ‘Neural Networks based Conformal Prediction for Pipeline Structural Response’, in Conformal and Probabilistic Prediction with Applications, 13-15 September 2023, Limassol, Cyprus. Volume 204 of Proceedings of Machine Learning Research, pp. 134-146. PMLR.
Hafver, A., Ferreira, C., Agrell, C., McGeorge, D., Hektor, E.A., Pedersen, F.B., van der Meulen, M., Haugen, O.I., Eldevik, S. and Myhrvold, T. (2021) ‘On the meaning of assurance’, in 31st European Safety and Reliability Conference.
