Neurosymbolic Object-Centric Learning with Distant Supervision

2025-06-19

AI Summary

This work addresses the challenge of unsupervised learning of object-centric representations directly from raw, unstructured perceptual data. We propose the first end-to-end differentiable neuro-symbolic framework that integrates probabilistic logic programming (ProbLog) with neural networks. Our method requires only task-level distant supervision—no object annotations, predefined segmentation masks, or explicit object priors. By jointly optimizing a probabilistic logic reasoning layer with a visual perception module, the model autonomously discovers semantically coherent object structures. Empirically, it significantly outperforms both pure neural networks and existing neuro-symbolic approaches on challenging benchmarks involving compositional object generalization, zero-shot task transfer, and variable numbers of objects. To our knowledge, this is the first approach to enable differentiable, logic-guided learning of object-centric representations under distant supervision.

Technology Category

• Machine Learning: Neuro-Symbolic Learning
• Computer Vision: Visual Reasoning & Symbolic Representations
• Knowledge Representation and Reasoning: Logic Programming

Application Category

• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Relational learning enables models to generalize across structured domains by reasoning over objects and their interactions. While recent advances in neurosymbolic reasoning and object-centric learning bring us closer to this goal, existing systems rely either on object-level supervision or on a predefined decomposition of the input into objects. In this work, we propose a neurosymbolic formulation for learning object-centric representations directly from raw unstructured perceptual data and using only distant supervision. We instantiate this approach in DeepObjectLog, a neurosymbolic model that integrates a perceptual module, which extracts relevant object representations, with a symbolic reasoning layer based on probabilistic logic programming. By enabling sound probabilistic logical inference, the symbolic component introduces a novel learning signal that further guides the discovery of meaningful objects in the input. We evaluate our model across a diverse range of generalization settings, including unseen object compositions, unseen tasks, and unseen number of objects. Experimental results show that our method outperforms neural and neurosymbolic baselines across the tested settings.

Problem

• Learning object-centric representations from raw unstructured data
• Using distant supervision for neurosymbolic object-centric learning
• Enhancing generalization across unseen object compositions and tasks

Innovation

• Neurosymbolic learning from raw unstructured data
• Distant supervision without object-level labels
• Probabilistic logic programming for object discovery