Scholar
Hector Geffner
Google Scholar ID: Wd0CDmcAAAAJ
RWTH Aachen University
Citations & Impact
Citations
4,478
H-index
37
i10-index
82
Publications
20
Co-authors
32
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Published multiple papers, including:
- - Learning Lifted STRIPS Models from Action Traces Alone: A Simple, General, and Scalable Solution (ICAPS 2025)
- - Learning More Expressive General Policies for Classical Planning Domains (AAAI 2025)
- - Learning Sketch Decompositions in Planning via Deep Reinforcement Learning (IJCAI 2025)
- - Learning to Ground Existentially Quantified Goals (KR 2024)
- - Symmetries and Expressive Requirements for Learning General Policies (KR 2024)
- - General Policies, Subgoal Structure, and Planning Width (JAIR 2024)
- - Expressing and Exploiting Subgoal Structure in Classical Planning Using Sketches (JAIR 2024)
- - On Policy Reuse: A Language for Representing and Executing Nested Policies (ICAPS 2024)
- - Combined Task and Motion Planning Via Sketch Decompositions (ICAPS 2024)
- - Learning Generalized Policies for Fully Observable Non-Deterministic Planning Domains (IJCAI 2024)
- - Equivalence-Based Abstractions for Learning General Policies (ICAPS Workshop (PRL) 2024)
- - General and Reusable Indexical Policies and Sketches (NeurIPS Workshop (GenPlan) 2023)
- - Learning Hierarchical Policies by Iteratively Reducing the Width of Sketch Rules (KR 2023)
- - Learning Generalized Policies with Policy Gradient Methods (KR 2023)
- - Combined Task and Motion Planning Via Sketch Decompositions (ICAPS PlanRob Workshop 2023)
- - Learning Generalized Policies Without Supervision Using GNNs (KR 2022)
- - Learning First-Order Symbolic Planning Representations That Are Grounded (ICAPS PRL Workshop, 2022)
Research Experience
- Leads a research group on learning to act and plan, funded by the ERC, the Humboldt Foundation, and RWTH Aachen. There are openings for highly motivated and talented postdocs, but not accepting new PhD students or interns.
Background
- Alexander von Humboldt Professor at the Computer Science Department, RWTH Aachen University, Germany, where he heads the Chair of Machine Learning and Reasoning.
Miscellany
- Organizes an annual Aachen Symposium on Representation learning to act and plan (RLeap).
