Bibliographic Details
| Title: |
Sparse Relational Reasoning with Object-Centric Representations |
| Authors: |
Spies, Alex F., Russo, Alessandra, Shanahan, Murray |
| Publication Year: |
2022 |
| Subject Terms: |
Computer Science - Machine Learning, Computer Science - Artificial Intelligence, I.2.10, I.2.6 |
| More Details: |
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.
Comment: ICML 2022, DyNN Workshop |
| Document Type: |
Working Paper |
| Access URL: |
http://arxiv.org/abs/2207.07512 |
| Accession Number: |
edsarx.2207.07512 |
| Database: |
arXiv |
