Scholar
Hidde Fokkema
Google Scholar ID: FkAOYFsAAAAJ
PhD student, University of Amsterdam
Citations & Impact
Citations
52
H-index
3
i10-index
1
Publications
6
Co-authors
11
list available
Publications
Resume (English only)
Academic Achievements
- Paper 'Attribution-based Explanations that Provide Recourse Cannot be Robust' published in Journal of Machine Learning Research (JMLR), 2023
- Paper 'Online Newton Method for Bandit Convex Optimisation' accepted at COLT 2024
- Paper accepted at AISTATS (Conference on AI and Statistics), 2024
- Two papers ('Concept Paper' and 'Performative Validity Paper') accepted at NeurIPS 2025
- Preprint 'Performative Validity of Recourse Explanations' released in June 2025
- Preprint 'Sample-efficient Learning of Concepts with Theoretical Guarantees: from Data to Concepts without Interventions' released in February 2025
- Preprint 'Risks of Recourse in Binary Classification' released in May 2023
- Presented research at international venues including ICML, COLT, AISTATS, JMLR-to-conference track, and workshops on interpretable machine learning
Background
- PhD Candidate in Mathematical Machine Learning at the Korteweg-de Vries Institute, University of Amsterdam
- Supervised by Dr. Tim van Erven
- Broad interest in Mathematics and Machine Learning
- Current research focuses on formal approaches to Explainable AI
- Other interests include: mathematical foundations of machine learning, (Martingale/Causal) Optimal Transport, and Stochastic Calculus
