Solving Raven's Progressive Matrices with Multi-Layer Relation Networks

Marius Jahrens, Thomas Martinetz

Abstract

Raven's Progressive Matrices are a benchmark originally designed to test the cognitive abilities of humans. It has recently been adapted to test relational reasoning in machine learning systems. For this purpose the so-called Procedurally Generated Matrices dataset was set up, which is so far one of the most difficult relational reasoning benchmarks. Here we show that deep neural networks are capable of solving this benchmark, reaching an accuracy of 98.0 percent over the previous state-of-the-art of 62.6 percent by combining Wild Relation Networks with Multi-Layer Relation Networks and introducing Magnitude Encoding, an encoding scheme designed for late fusion architectures.

Original languageEnglish
Title of host publication2020 International Joint Conference on Neural Networks (IJCNN)
PublisherIEEE
Publication date07.2020
Article number9207319
ISBN (Print)978-1-7281-6927-9
ISBN (Electronic)978-1-7281-6926-2
DOIs
Publication statusPublished - 07.2020
Event2020 International Joint Conference on Neural Networks - Virtual, Glasgow, United Kingdom
Duration: 19.07.202024.07.2020
Conference number: 163566

Research Areas and Centers

  • Research Area: Intelligent Systems
  • Centers: Center for Artificial Intelligence Luebeck (ZKIL)

DFG Research Classification Scheme

  • 409-05 Interactive and Intelligent Systems, Image and Language Processing, Computer Graphics and Visualisation

Fingerprint

Dive into the research topics of 'Solving Raven's Progressive Matrices with Multi-Layer Relation Networks'. Together they form a unique fingerprint.

Cite this