An Explicitly Relational Neural Network Architecture
arXiv:1905.10307
Abstract
With a view to bridging the gap between deep learning and symbolic AI, we present a novel end-to-end neural network architecture that learns to form propositional representations with an explicitly relational structure from raw pixel data. In order to evaluate and analyse the architecture, we introduce a family of simple visual relational reasoning tasks of varying complexity. We show that the proposed architecture, when pre-trained on a curriculum of such tasks, learns to generate reusable representations that better facilitate subsequent learning on previously unseen tasks when compared to a number of baseline architectures. The workings of a successfully trained model are visualised to shed some light on how the architecture functions.
In Proceedings ICML 2020
References in corpus (7)
- Interaction Networks for Learning about Objects, Relations and Physics
- Deep Learning: A Critical Appraisal
- Universal Transformers
- Schema Networks: Zero-shot Transfer with a Generative Causal Model of Intuitive Physics
- Towards Deep Symbolic Reinforcement Learning
- SCAN: Learning Hierarchical Compositional Visual Concepts
- Unsupervised Grounding of Plannable First-Order Logic Representation from Images
Cited by in corpus (7)
- A Comprehensive Survey of Machine Learning Applied to Radar Signal Processing
- Structure in Deep Reinforcement Learning: A Survey and Open Problems
- Emergent Symbols through Binding in External Memory
- An Explicit Local and Global Representation Disentanglement Framework with Applications in Deep Clustering and Unsupervised Object Detection
- Unifying Neural Learning and Symbolic Reasoning for Spinal Medical Report Generation
- A Study of Compositional Generalization in Neural Models
- pix2rule: End-to-end Neuro-symbolic Rule Learning