Scholar
Sascha Marton
Google Scholar ID: 5PQJ3sEAAAAJ
University of Mannheim
Citations & Impact
Citations
122
H-index
6
i10-index
5
Publications
16
Co-authors
5
list available
Publications
Resume (English only)
Academic Achievements
- - Publications:
- - Mitigating Information Loss in Tree-Based Reinforcement Learning via Direct Optimization, ICLR 2025 (Spotlight)
- - Decision Trees That Remember: Gradient-Based Learning of Recurrent Decision Trees with Memory, ICLR 2025 NFAM Workshop
- - GRANDE: Gradient-Based Decision Tree Ensembles for Tabular Data, ICLR 2024
- - GradTree: Learning Axis-Aligned Decision Trees with Gradient Descent, AAAI 2024 (Oral)
- - Explaining neural networks without access to training data, Machine Learning Journal (2024)
Research Experience
- - Assistant Professor (Akademischer Rat), Technical University of Clausthal, Germany
- - Research Projects: Developing a new approach for learning hard, axis-aligned decision trees using gradient descent; proposing ReMeDe trees, a recurrent decision tree architecture with internal memory, enabling efficient learning for sequential data.
Education
- - PhD in Machine Learning, University of Mannheim, May 2025, Advisor: Heiner Stuckenschmidt
Background
- - Research Interests: Ensemble Methods, Tree-Based Methods, Deep Learning for Tabular Data, Time-Series Forecasting, Explainable Artificial Intelligence
- - Professional Field: Machine Learning
- - Brief Introduction: Assistant Professor (Akademischer Rat) at the Technical University of Clausthal in Germany, focusing on advancing machine learning techniques for tabular data, particularly gradient-based decision tree learning and tree-based ensemble methods.
