Jannik Kossen
Papers
1
Total Citations
2
H-Index
1
About
Jannik Kossen is a researcher whose work sits at the intersection of machine learning, decision-making, and data efficiency. His primary research areas include active learning, Bayesian optimization, and uncertainty quantification, with a particular focus on developing algorithms that can intelligently acquire costly or scarce data. Kossen’s major contribution is the formulation of the "Active Acquisition for Multimodal Temporal Data" (A2MT) task, a challenging decision-making framework that addresses real-world scenarios where input features are not freely available at test time but must be procured at significant expense. This work, published in 2022, has garnered early attention with 2 citations, reflecting its foundational nature in a nascent but critical area. By designing agents that actively decide which data to acquire, Kossen’s research promises to reduce costs and improve efficiency in applications ranging from medical diagnostics to autonomous systems. His work stands out for its practical relevance, bridging the gap between theoretical active learning and the complexities of multimodal, time-dependent data. As a rising voice in this field, Kossen’s contributions are poised to influence how machines learn under real-world resource constraints.
Research Focus
Key Achievements
Top Papers
- 1