A simple neural network module for relational reasoning

Relational reasoning is a central component of generally intelligent behavior, but has proven difficult for neural networks to learn. In this paper we describe how to use Relation Networks (RNs) as a simple plug-and-play module to solve problems that fundamentally hinge on relational reasoning... (read more)

PDF Abstract NeurIPS 2017 PDF NeurIPS 2017 Abstract

Datasets


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Visual Question Answering CLEVR CNN + LSTM + RN Accuracy 95.50 # 13
Image Retrieval with Multi-Modal Query Fashion200k Relationship Recall@1 13 # 3
Recall@10 40.5 # 3
Recall@50 62.4 # 3

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet