Junior Research Group Leader (m/f/d) (DM-614)

Overview

The Berlin Mathematics Research Center MATH+ is a dynamic interdisciplinary research center that focuses on application-driven mathematical research relevant to our society. Its three main goals are to train highly skilled young mathematicians across the entire breadth of mathematics, to develop innovative application-oriented mathematics with a focus on data-driven modelling, simulation, and optimization, and to open up new mathematical thinking spaces. MATH+ is supported by Berlin mathematics as a whole in cooperation with other scientific disciplines. Moreover, MATH+ strives to communicate its findings actively to the general public.

What You Can Expect In Your New Role

The Department of Mathematics and Computer Science at Freie Universität Berlin invites applications for an independent Junior Research Group Leader in Statistical Learning. The successful candidate will establish and lead a research group developing novel mathematical and statistical methodologies for modern data analysis. The position offers a unique opportunity to pursue foundational research at the interface of statistics, machine learning, and applied mathematics. Within the framework of the MATH+ Cluster of Excellence, the group will contribute to advancing the theoretical understanding of learning from complex, structured, or high-dimensional data, and to exploring principled methods for data-driven modeling, inference, and prediction.

We Encourage Candidates Pursuing Innovative Research In Theoretical And Mathematical Aspects Of Statistical Learning And Modern Data Science. Topics Of Interest Within The Expertise Of The Candidate Include (but Are Not Limited To)

• Foundations of machine and deep learning from a mathematical and statistical perspective, including generalization, optimization, and expressivity
• Regularization, sparsity, and model selection in high-dimensional statistics
• Bayesian and probabilistic approaches to statistical inference and learning
• Reinforcement and sequential decision learning
• Generative and probabilistic modeling, such as diffusion or flow-based methods for data-driven inference
• Connections to optimal transport, including theoretical and computational aspects for data-driven modeling and learning
• Learning from complex models described by ODEs, PDEs, or SDEs, connecting to areas such as inverse problems, uncertainty quantification, and stochastic dynamics
• Learning in complex systems, including multi-agent dynamics, societal processes, opinion formation, or climate impact modeling

An interim evaluation will take place within the Cluster of Excellence framework after three years.

Key Requirements

Doctoral degree in mathematics, statistics, computer science, or a related discipline.

Desirable

• Several years of postdoctoral experience with demonstrated independence
• Proven research record in statistical learning, machine learning, or high dimensional statistics
• Strong related publication record
• Previous experience on attracting third-party funding
• Excellent communication and collaboration skills across disciplines
• Gender and diversity competence

Benefits And Other Advantages

A start-up package, dedicated office, shared facilities (data management, computing, etc.) and administrative support will be provided.

Application Process

If you are interested in what we have to offer, you may send your application materials directly to us. We expect to receive documents such as your CV, certificates, a list of your publications and references. You should also provide a brief description of your research field, your achievements and goals to date, and your vision for yourself and your (future) research group. Your application should clearly demonstrate your motivation for the position and your profile for yourself and the research group. Simply submit your application via the Online Recruiting Portal by clicking the blue “Apply Now” button. From there you will be redirected to set up a profile (only necessary for your first application).

You can also get in touch with Prof. Schillings ( ).

#J-18808-Ljbffr