cv

PhD Candidate

Bielefeld University

Researching on Explainable AI and Game Theory

SAS & Data Science Consultant

Positive Thinking Company (former mayato)

Consulting International Clients in Banking and Insurance

SAS & Data Science Consultant

mayato

Consulting Clients in Berlin Area

Data Scientist & Machine Learning Engineer

Munich Re

Machine Learning Solutions for Health Insurance

Teaching Assistant

Freie Universität Berlin

Tutorial Lectures and Exam Grading

R & Web Developer

ISQ Berlin Brandenburg e.V.

Voluntary Trainer & Coordinator

TuSLi Berlin (Table Tennis)

Leading Training Sessions, Coaching Competitions & Organization

• Acquired DOSB Licence C

PhD

Bielefeld University, Bielefeld, Germany

Machine Learning (Computer Science)

• Explainable AI

Erasmus+

Norges teknisk-naturvitenskapelige universitet (NTNU), Norway

Statistics

MSc.

Technische Universität Berlin, Berlin, Germany

Mathematics

• Stochastics & Financial Mathematics
• Mathematical Deep Learning

BSc.

Freie Universität Berlin, Berlin, Germany

Mathematics

• Stochastics & Financial Mathematics
• Mathematical Deep Learning

Best Paper Award

World Conference on eXplainable Artificial Intelligence (xAI 2023)

iPDP: On Partial Dependence Plots in Dynamic Modeling Scenarios

Best Paper in Journal Track

European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD 2023)

Incremental Permutation Feature Importance (iPFI): Towards Online Explanations on Data Streams

Top Reviewer

Neural Information Processing Systems (NeurIPS 2024)

Best Reviewer

International Conference on Machine Learning (ICML 2024)

Unifying Feature-Based Explanations with Functional ANOVA and Cooperative Game Theory

Preprint

This paper presents a unifying framework for feature-based explanations.

shapiq: Shapley Interactions for Machine Learning

NeurIPS 2024

This package paper presents Shapley interactions for 11 machine learning domains, including an extension of the popular shap package with feature interactions.

KernelSHAP-IQ: Weighted Least Square Optimization for Shapley Interactions

ICML 2024

We present KernelSHAP for Shapley Interactions and a weighted least-squares representation of the Shapley interaction index.

Beyond TreeSHAP:Efficient Computation of Any-Order Shapley Interactions for Tree Ensembles

AAAI 2024

We present TreeSHAP for Shapley interactions that works for any tree-based model.

SVARM-IQ: Efficient Approximation of Any-order Shapley Interactions through Stratification

AISTATS 2024

We present SVARM for Shapley interactions by using stratification.

iPDP: On Partial Dependence Plots in Dynamic Modeling Scenarios

World Conference on eXplainable Artificial Intelligence (xAI 2023)

We present computation of PDPs efficiently in dynamically evolving environments, such as online learning on data streams.

SHAP-IQ: Unified Approximation of any-order Shapley Interactions

NeurIPS 2023

We present a unified approximation method for Shapley interactions.

Incremental permutation feature importance (iPFI): towards online explanations on data streams

Machine Learning Journal

We present an efficient computation of PFI on evolving data streams with important theoretical guarantees.

iSAGE: An Incremental Version of SAGE for Online Explanation on Data Streams

ECML PKDD 2023

We present an efficient computation of SAGE on evolving data streams.

shapiq

Shapley interactions for Machine Learning & Extension of shap

• Feature Interactions
• Shapley Interactions
• Game Theory for Machine Learning