αILP: thinking visual scenes as differentiable logic programs

Deep neural learning has shown remarkable performance at learning representations for visual object categorization. However, deep neural networks such as CNNs do not explicitly encode objects and relations among them. This limits their success on tasks that require a deep logical understanding of vi...

Full description

Saved in:
Bibliographic Details
Published in:Machine learning Vol. 112; no. 5; pp. 1465 - 1497
Main Authors: Shindo, Hikaru, Pfanschilling, Viktor, Dhami, Devendra Singh, Kersting, Kristian
Format: Journal Article
Language:English
Published: New York Springer US 01.05.2023
Springer Nature B.V
Subjects:
Artificial Intelligence
Artificial neural networks
Computer Science
Confounding (Statistics)
Control
Logic programming
Logic programs
Machine Learning
Mechatronics
Natural Language Processing (NLP)
Neural networks
Robotics
Simulation and Modeling
Special Issue on Learning and Reasoning 2022
Visual perception
Object-centric learning
Neuro-symbolic AI
Inductive logic programming
Differentiable reasoning
ISSN:0885-6125, 1573-0565
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Deep neural learning has shown remarkable performance at learning representations for visual object categorization. However, deep neural networks such as CNNs do not explicitly encode objects and relations among them. This limits their success on tasks that require a deep logical understanding of visual scenes, such as Kandinsky patterns and Bongard problems. To overcome these limitations, we introduce α ILP , a novel differentiable inductive logic programming framework that learns to represent scenes as logic programs—intuitively, logical atoms correspond to objects, attributes, and relations, and clauses encode high-level scene information. α ILP has an end-to-end reasoning architecture from visual inputs. Using it, α ILP performs differentiable inductive logic programming on complex visual scenes, i.e., the logical rules are learned by gradient descent. Our extensive experiments on Kandinsky patterns and CLEVR-Hans benchmarks demonstrate the accuracy and efficiency of α ILP in learning complex visual-logical concepts.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0885-6125
1573-0565
DOI:10.1007/s10994-023-06320-1