Scholar
Hikaru Shindo
Google Scholar ID: Ws03zBoAAAAJ
TU Darmstadt
Citations & Impact
Citations
222
H-index
7
i10-index
5
Publications
20
Co-authors
18
list available
Contact
Publications
Resume (English only)
Academic Achievements
- 2025: BlendRL: A Framework for Merging Symbolic and Neural Policy Learning. ICLR 2025, Spotlight.
- 2025: ART: Adaptive Relation Tuning for Generalized Relation Detection. ICCV 2025.
- 2025: Gestalt Vision: A Dataset for Evaluating Gestalt Principles in Visual Perception. NeSy 2025, PMLR.
- 2025: NeST: The Neuro-Symbolic Transpiler. IJAR.
- 2025: Human-Allied Relational Reinforcement Learning. ACS 2025, Oral.
- 2024: DeiSAM: Segment Anything with Deictic Prompting. NeurIPS 2024.
- 2024: Learning Differentiable Logic Programs for Abstract Visual Reasoning. MLJ.
- 2024: Neural-Symbolic Predicate Invention: Learning Relational Concepts from Visual Scenes. NAIJ.
- 2024: Learning by Self-Explaining. TMLR.
- 2024: V-LoL: A Diagnostic Dataset for Visual Logical Learning. DMLR.
- 2023: alphaILP: Thinking Visual Scenes as Differentiable Logic Programs. MLJ.
- 2023: Interpretable and Explainable Logical Policies via Neurally Guided Symbolic Abstraction. NeurIPS 2023.
Research Experience
- Building differentiable reasoning pipelines and applying them to practical tasks of visual understanding and reinforcement learning.
Education
- Ph.D. candidate, AI/ML group, TU Darmstadt, Germany
Background
- Research interests include developing intelligent agents that can perceive, reason, and act in complex environments. Focuses on neuro-symbolic methods, combining neural networks with symbolic reasoning to enable both low-level perception and high-level decision-making.
Miscellany
- Contact: hikaru.shindo@tu-darmstadt.de
