Abstract Reasoning with Distracting Features

12/02/2019
by   Kecheng Zheng, et al.
16

Abstraction reasoning is a long-standing challenge in artificial intelligence. Recent studies suggest that many of the deep architectures that have triumphed over other domains failed to work well in abstract reasoning. In this paper, we first illustrate that one of the main challenges in such a reasoning task is the presence of distracting features, which requires the learning algorithm to leverage counterevidence and to reject any of the false hypotheses in order to learn the true patterns. We later show that carefully designed learning trajectory over different categories of training data can effectively boost learning performance by mitigating the impacts of distracting features. Inspired by this fact, we propose feature robust abstract reasoning (FRAR) model, which consists of a reinforcement learning based teacher network to determine the sequence of training and a student network for predictions. Experimental results demonstrated strong improvements over baseline algorithms and we are able to beat the state-of-the-art models by 18.7 dataset and 13.3

READ FULL TEXT

page 2

page 3

page 4

page 5

page 7

page 8

page 10

page 11

research
07/11/2018

Measuring abstract reasoning in neural networks

Whether neural networks can learn abstract reasoning or whether they mer...
research
01/31/2019

Learning to Make Analogies by Contrasting Abstract Relational Structure

Analogical reasoning has been a principal focus of various waves of AI r...
research
10/12/2021

AVoE: A Synthetic 3D Dataset on Understanding Violation of Expectation for Artificial Cognition

Recent work in cognitive reasoning and computer vision has engendered an...
research
12/03/2021

An Impossible Asylum

In 1982, Raymond Smullyan published an article, "The Asylum of Doctor Ta...
research
11/12/2020

Neural Abstract Reasoner

Abstract reasoning and logic inference are difficult problems for neural...
research
06/14/2015

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...
research
10/18/2022

Graphs, Constraints, and Search for the Abstraction and Reasoning Corpus

The Abstraction and Reasoning Corpus (ARC) aims at benchmarking the perf...

Please sign up or login with your details

Forgot password? Click here to reset