paper-with-me

홈 › Papers

Sparse Relational Reasoning with Object-Centric Representations

2022-07-15 · Alex F. Spies, Alessandra Russo, Murray Shanahan

We investigate the composability of soft-rules learned by relational neural architectures when operating over object-centric (slot-based) representations, under a variety of sparsity-inducing constraints. We find that increasing sparsity, especially on features, improves the performance of some models and leads to simpler relations. Additionally, we observe that object-centric representations can be detrimental when not all objects are fully captured; a failure mode to which CNNs are less prone. These findings demonstrate the trade-offs between interpretability and performance, even for models designed to tackle relational tasks.

📄 PDF Abstract BibTeX arXiv:2207.07512

Code (0)

등록된 구현이 없습니다.

Tasks

ObjectRelational Reasoning

Similar Papers 제목 키워드 기반

Edge-Centric Relational Reasoning for 3D Scene Graph Prediction

2025-11-19 · Yanni Ma, Hao Liu, Yulan Guo, Theo Gevers 외 arxiv

3D scene graph prediction aims to abstract complex 3D environments into structured graphs consisting of objects and their pairwise relationships. Existing approaches typically adopt object-centric graph neural networks, …

Relational ReasoningGraph Neural Network

Systematic Visual Reasoning through Object-Centric Relational Abstraction

2023-06-04 · NeurIPS 2023 11 · Taylor W. Webb, Shanka Subhra Mondal, Jonathan D. Cohen

Human visual reasoning is characterized by an ability to identify abstract patterns from only a small number of examples, and to systematically generalize those patterns to novel inputs. This capacity depends in large pa…

ObjectSystematic GeneralizationVisual Reasoning

Agent-Centric Representations for Multi-Agent Reinforcement Learning

2021-04-19 · Wenling Shang, Lasse Espeholt, Anton Raichuk, Tim Salimans

Object-centric representations have recently enabled significant progress in tackling relational reasoning tasks. By building a strong object-centric inductive bias into neural architectures, recent efforts have improved…

Inductive BiasMulti-agent Reinforcement LearningObjectreinforcement-learning+4

OCRT: Boosting Foundation Models in the Open World with Object-Concept-Relation Triad

2025-03-24 · CVPR 2025 1 · Luyao Tang, Yuxuan Yuan, Chaoqi Chen, Zeyu Zhang 외

Although foundation models (FMs) claim to be powerful, their generalization ability significantly decreases when faced with distribution shifts, weak supervision, or malicious attacks in the open world. On the other hand…

Domain GeneralizationRelationRelational Reasoning

Relational inductive bias for physical construction in humans and machines

2018-06-04 · Jessica B. Hamrick, Kelsey R. Allen, Victor Bapst, Tina Zhu 외

While current deep learning systems excel at tasks such as object classification, language processing, and gameplay, few can construct or modify a complex system such as a tower of blocks. We hypothesize that what these …

Deep Reinforcement LearningInductive BiasObjectReinforcement Learning