I am a postdoctoral research and teaching fellow (French: ATER) in Computer Science at LIP6, Sorbonne University in Paris.
Previously, I obtained a PhD in Computer Science from the same university, funded by the EDITE doctoral scholarship, with Carola Doerr as my thesis supervisor, in December 2021. The manuscript of my thesis is available here.
My main research interests lie on the crossroads of black-box optimization and machine learning. In particular, I work on dynamic automated algorithm selection in numerical optimization, where the main challenge is to design an efficient and robust machine learning model that is able to track and adapt its recommendation of the best suited algorithm for a given problem instance throughout the optimization process. Our novel approach in this context is based on extracting the knowledge about problem instances by leveraging the points from the algorithm's search trajectory, as seen during optimization. This allows for a significant decrease in the overhead cost induced by the landscape feature extraction procedure, which is typically performed as a pre-processing step. For more in-depth information, please refer to my Publications page.
I am currently looking for a PostDoc position from September 2022 onwards!
© Anja Jankovic. All Rights Reserved. Designed by HTML Codex