CV

Education

• Ph.D in Machine Learning, University of Mannheim (summa cum laude, May 2025)
  • Dissertation: „Learning Tree-Based Models with Gradient Descent”
• M.S. in Business Informatics (Data Science Track), University of Mannheim, 2019
• B.S. in Business Informatics, University of Mannheim, 2017

Work experience

• 05/2025-now: Assistant Professor (Akademischer Rat)
  • Technical University of Clausthal, Germany
  • Topics: Deep Learning for Tabular Data, Tree Ensemble Methods, Concept Bottleneck Models, Reinforcement Learning with Verifiable Rewards (RLVR)
• 10/2024 - 03/2025: Adjunct Lecturer
  • Technical University of Clausthal, Germany
  • Lecture: “Grundlagen der Künstlichen Intelligenz” (Introduction to AI)
• 09/2019-05/2025: Scientific Researcher and PhD Candidate
  • University of Mannheim, Germany
  • Topics: Deep Learning for Tabular Data, Tree Ensemble Methods, Explainable AI

Skills

• Technical Skills
  • Programming (Python, R, Java (very good), C/C++ (basic)
  • ​​​​Frameworks & Tools: TensorFlow, PyTorch, JAX
• Research Areas:
  • Tabular Data
  • Time Series Forecasting
  • Reinforcement Learning
  • Concept Bottleneck Models
• Languages
  • German (native)
  • English (Fluent)
  • Italian (Conversational)

Publication List

Selected Publications

Mitigating Information Loss in Tree-Based Reinforcement Learning via Direct Optimization
Sascha Marton, Tim Grams, Florian Vogt, Stefan Lüdtke, Christian Bartelt, Heiner Stuckenschmidt
ICLR 2025 (Spotlight)

DCBM: Data-Efficient Visual Concept Bottleneck Models
Katharina Prasse, Patrick Knab, Sascha Marton, Christian Bartelt, Margret Keuper
ICML 2025

GRANDE: Gradient-Based Decision Tree Ensembles for Tabular Data
Sascha Marton, Stefan Lüdtke, Christian Bartelt, Heiner Stuckenschmidt
ICLR 2024

A Data-Centric Perspective on Evaluating Machine Learning Models for Tabular Data
Andrej Tschalzev, Sascha Marton, Stefan Lüdtke, Christian Bartelt, Heiner Stuckenschmidt
NeurIPS 2024 (Datasets and Benchmarks Track)

GradTree: Learning Axis-Aligned Decision Trees with Gradient Descent
Sascha Marton, Stefan Lüdtke, Christian Bartelt, Heiner Stuckenschmidt
AAAI 2024 (Oral)

Further Publications

Decision Trees That Remember: Gradient-Based Learning of Recurrent Decision Trees with Memory
Sascha Marton, Moritz Schneider, Jannik Brinkmann, Stefan Ludtke, Christian Bartelt, Heiner Stuckenschmidt
ICLR 2025 Workshop on New Frontiers in Associative Memories

Beyond Pixels: Enhancing LIME with Hierarchical Features and Segmentation Foundation Models
Patrick Knab, Sascha Marton, Christian Bartelt
ECAI 2025

Disentangling Exploration of Large Language Models by Optimal Exploitation
Tim Grams, Patrick Betz, Sascha Marton, Stefan Lüdtke and Christian Bartelt
ECAI 2025

Which LIME should I trust? Concepts, Challenges, and Solutions
Patrick Knab, Sascha Marton, Udo Schlegel, Christian Bartelt
XAI 2025

Explaining neural networks without access to training data
Sascha Marton, Stefan Lüdtke, Christian Bartelt, Heiner Stuckenschmidt
Machine Learning Journal (2024)

Explanations for Neural Networks by Neural Network
Sascha Marton, Stefan Lüdtke, Christian Bartelt
Applied Sciences (2022)

Interpreting Outliers in Time Series Data through Decoding Autoencoder
Patrick Knab, Sascha Marton, Christian Bartelt, Robert Fuder
ECML-PKDD 2024 Workshop on Explainable AI for Time Series and Data Streams

Bias mitigation for large language models using adversarial learning
Jasmina S Ernst, Sascha Marton, Jannik Brinkmann, Eduardo Vellasques, Damien Foucard, Martin Kraemer, Marian Lambert
ECAI 2023 Workshop on Fairness and Bias in AI