We’re happy to share that Maryam Gholami Shiri presented a paper titled, “Design Choices That Matter: A Functional ANOVA Analysis for Remote Sensing Multi-Label Classification”, at Discovery Science 2026 in Mainz, Germany!

AutoLearn-SI work uses functional analysis of variance (fANOVA) to move beyond model rankings and examine how design choices—and their interactions—contribute to performance variability. Understanding these effects can help researchers make more informed decisions when developing models for remote sensing tasks.

Congratulations to Maryam Gholami Shiri on the presentation, and thank you to the co-authors Eva Tuba, Saso Dzeroski, Tome Eftimov, and Ana Nikolikj!