Jakob Gawlikowski

Deutsches Zentrum für Luft- und Raumfahrt e. V. (DLR)

Papers

Total Citations

1,134

H-Index

About

Jakob Gawlikowski is a leading researcher at the intersection of deep learning and uncertainty quantification, a field critical for building trustworthy AI systems. His most influential work, the comprehensive 2023 survey "A survey of uncertainty in deep neural networks," has already amassed over 1,100 citations, establishing itself as a go-to resource for understanding how neural networks can express confidence—or doubt—in their predictions. This foundational contribution systematically maps the landscape of Bayesian and non-Bayesian approaches, from Monte Carlo dropout to ensemble methods, making complex probabilistic concepts accessible to the broader machine learning community. Gawlikowski’s research addresses a fundamental challenge: while modern neural networks excel at pattern recognition, they often fail to signal when they are uncertain, a critical flaw for high-stakes applications in autonomous driving, medical imaging, and climate modeling. By advancing methods that allow models to “know what they don’t know,” his work directly enhances the safety and reliability of AI deployments. His contributions are particularly valuable for students and practitioners seeking to build robust systems that can gracefully handle out-of-distribution data and adversarial inputs.