Abstract Reasoning via Logic-guided Generation

by   Sihyun Yu, et al.

Abstract reasoning, i.e., inferring complicated patterns from given observations, is a central building block of artificial general intelligence. While humans find the answer by either eliminating wrong candidates or first constructing the answer, prior deep neural network (DNN)-based methods focus on the former discriminative approach. This paper aims to design a framework for the latter approach and bridge the gap between artificial and human intelligence. To this end, we propose logic-guided generation (LoGe), a novel generative DNN framework that reduces abstract reasoning as an optimization problem in propositional logic. LoGe is composed of three steps: extract propositional variables from images, reason the answer variables with a logic layer, and reconstruct the answer image from the variables. We demonstrate that LoGe outperforms the black box DNN frameworks for generative abstract reasoning under the RAVEN benchmark, i.e., reconstructing answers based on capturing correct rules of various attributes from observations.


Logic could be learned from images

Logic reasoning is a significant ability of human intelligence and also ...

Reasoning about Minimal Belief and Negation as Failure

We investigate the problem of reasoning in the propositional fragment of...

Abstract Learning via Demodulation in a Deep Neural Network

Inspired by the brain, deep neural networks (DNN) are thought to learn a...

Artificial general intelligence through recursive data compression and grounded reasoning: a position paper

This paper presents a tentative outline for the construction of an artif...

Raven's Progressive Matrices Completion with Latent Gaussian Process Priors

Abstract reasoning ability is fundamental to human intelligence. It enab...

Abstract Modular Systems and Solvers

Integrating diverse formalisms into modular knowledge representation sys...

Bilingual analogical proportions

Analogical proportions are expressions of the form “a is to b what c is ...

Please sign up or login with your details

Forgot password? Click here to reset