Feature-Based Explainable AI: From Attribution to Interaction, From Static to Dynamic. Dissertation

MCML Authors

Maximilian Muschalik

Dr.

→ Group Eyke Hüllermeier
 Artificial Intelligence and Machine Learning

Abstract

This dissertation advances explainable AI for complex and evolving machine learning models by developing efficient methods to explain both individual feature contributions and higher-order feature interactions. It also introduces incremental explanation techniques for data stream settings, enabling explanations to remain accurate as models continuously learn from new data. (Shortened.)

phdthesis Mus26

---

Dissertation

LMU München. Jun. 2026

---

Authors

M. Muschalik

---

Links

DOI

---

Research Area

 A3 | Computational Models