Scale-Localized Abstract Reasoning

09/20/2020
by   Yaniv Benny, et al.
0

We consider the abstract relational reasoning task, which is commonly used as an intelligence test. Since some patterns have spatial rationales, while others are only semantic, we propose a multi-scale architecture that processes each query in multiple resolutions. We show that indeed different rules are solved by different resolutions and a combined multi-scale approach outperforms the existing state of the art in this task on all benchmarks by 5-54 of our method is shown to arise from multiple novelties. First, it searches for relational patterns in multiple resolutions, which allows it to readily detect visual relations, such as location, in higher resolution, while allowing the lower resolution module to focus on semantic relations, such as shape type. Second, we optimize the reasoning network of each resolution proportionally to its performance, hereby we motivate each resolution to specialize on the rules for which it performs better than the others and ignore cases that are already solved by the other resolutions. Third, we propose a new way to pool information along the rows and the columns of the illustration-grid of the query. Our work also analyses the existing benchmarks, demonstrating that the RAVEN dataset selects the negative examples in a way that is easily exploited. We, therefore, propose a modified version of the RAVEN dataset, named RAVEN-FAIR. Our code and pretrained models are available at https://github.com/yanivbenny/MRNet. The dataset of RAVEN-FAIR is available at https://github.com/yanivbenny/RAVEN_FAIR.

READ FULL TEXT
research
02/18/2022

MultiRes-NetVLAD: Augmenting Place Recognition Training with Low-Resolution Imagery

Visual Place Recognition (VPR) is a crucial component of 6-DoF localizat...
research
03/25/2021

USB: Universal-Scale Object Detection Benchmark

Benchmarks, such as COCO, play a crucial role in object detection. Howev...
research
08/04/2018

Learning Multi-scale Features for Foreground Segmentation

Foreground segmentation algorithms aim segmenting moving objects from th...
research
11/19/2019

HighEr-Resolution Network for Image Demosaicing and Enhancing

Neural-networks based image restoration methods tend to use low-resoluti...
research
09/19/2017

A Fast and Accurate Vietnamese Word Segmenter

We propose a novel approach to Vietnamese word segmentation. Our approac...
research
06/13/2019

Cognitive Knowledge Graph Reasoning for One-shot Relational Learning

Inferring new facts from existing knowledge graphs (KG) with explainable...
research
07/19/2020

Resolution Switchable Networks for Runtime Efficient Image Recognition

We propose a general method to train a single convolutional neural netwo...

Please sign up or login with your details

Forgot password? Click here to reset