R-SQAIR: Relational Sequential Attend, Infer, Repeat

2019/10/11 by Aleksandar Stanić, Jürgen Schmidhuber, Stanić, Aleksandar +1 · 3 citations
Computer Science · #Bayesian Modeling and Causal Inference #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #I.2.6 #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Time Series Analysis and Forecasting

paper · pdf · doi:10.48550/arxiv.1910.05231

openalex publication_date 2019/10/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Traditional sequential multi-object attention models rely on a recurrent mechanism to infer object relations. We propose a relational extension (R-SQAIR) of one such attention model (SQAIR) by endowing it with a module with strong relational inductive bias that computes in parallel pairwise interactions between inferred objects. Two recently proposed relational modules are studied on tasks of unsupervised learning from videos. We demonstrate gains over sequential relational mechanisms, also in terms of combinatorial generalization.

Citations

• Semi-Supervised Learning with Ladder Networks
• Long Short-Term Memory
• Infants’ Metaphysics: The Case of Numerical Identity
• Recurrent World Models Facilitate Policy Evolution
• COBRA: Data-Efficient Model-Based RL through Unsupervised Object Discovery and Curiosity-Driven Exploration
• A Perspective on Objects and Systematic Generalization in Model-Based RL

Cited by

• SlotPi: Physics-informed Object-centric Reasoning Models
• Learning Physical Graph Representations from Visual Scenes
• Unsupervised Learning of Compositional Energy Concepts

Related

• SARN: Relational Reasoning through Sequential Attention
• Relational Embedding for Few-Shot Classification
• Abstractors and relational cross-attention: An inductive bias for explicit relational reasoning in Transformers
• Relational inductive biases on attention mechanisms
• Relational Neural Expectation Maximization: Unsupervised Discovery of Objects and their Interactions