A Perspective on Objects and Systematic Generalization in Model-Based RL

2019/06/03 by Sjoerd van Steenkiste, Klaus Greff, van Steenkiste, Sjoerd +3
Computer Science · #AI-based Problem Solving and Planning #Artificial Intelligence (cs.AI) #Constraint Satisfaction and Optimization #FOS: Computer and information sciences #I.2.6 #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Neural and Evolutionary Computing (cs.NE) #Semantic Web and Ontologies

paper · pdf · doi:10.48550/arxiv.1906.01035

openalex publication_date 2019/06/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

In order to meet the diverse challenges in solving many real-world problems, an intelligent agent has to be able to dynamically construct a model of its environment. Objects facilitate the modular reuse of prior knowledge and the combinatorial construction of such models. In this work, we argue that dynamically bound features (objects) do not simply emerge in connectionist models of the world. We identify several requirements that need to be fulfilled in overcoming this limitation and highlight corresponding inductive biases.

Citations

• Poincaré Embeddings for Learning Hierarchical Representations
• AI Safety Gridworlds
• Recurrent World Models Facilitate Policy Evolution
• Learning to Reason with Third-Order Tensor Products

Related

• Deep Sets for Generalization in RL
• General agents contain world models
• An Object Model for the Representation of Empirical Knowledge
• Feature-Attending Recurrent Modules for Generalization in Reinforcement Learning
• Supervised learning on heterogeneous, attributed entities interacting\n over time