Abstract: I will provide an overview of our new Virtual Laboratory for Molecular Level Atmospheric Transformations (VILMA) Centre of Excellence and discuss why explainable AI (XAI) and quantifying uncertainties is essential for us. As an example of our ongoing work, I will describe SLISEMAP, a supervised manifold visualisation method, a technique for XAI developed by us that finds local explanations for all data items. SLISEMAP produces (typically) two-dimensional global visualisation of the black box model such that data items with similar local explanations will be embedded nearby.

References :

· Virtual Laboratory for Molecular Level Atmospheric Transformations (VILMA) Centre of Excellence, https://www.helsinki.fi/en/researchgroups/vilma.

· Björklund, A., Mäkelä, J., & Puolamäki, K. (2022). SLISEMAP: Supervised dimensionality reduction through local explanations. Machine Learning. https://doi.org/10.1007/s10994-022-06261-1

Speakers:  Kai Puolamäki is an Associate Professor in computer science and atmospheric sciences at the Department of Computer Science and Institute for Atmospheric and Earth System Research at the University of Helsinki. He has a PhD in theoretical physics and holds a Title of Docent in computer science. His research interests include machine learning and exploratory data analysis. Kai Puolamäki has website at https://www.iki.fi/kaip.

Affiliation: University of Helsinki

Place of Seminar:  Kumpula exactum D122 (in person) & zoom ( Meeting ID: 640 5738 7231 ; Passcode: 825217)