A Cognitively-Inspired Neural Architecture for Visual Abstract Reasoning Using Contrastive Perceptual and Conceptual Processing
This work addresses the problem of improving machine performance on visual abstract reasoning tasks, such as intelligence tests, for researchers in AI and cognitive science, though it is incremental as it builds on existing datasets and methods.
The paper tackles visual abstract reasoning by introducing a cognitively-inspired neural architecture that iteratively contrasts perceptual and conceptual processing, achieving higher accuracy than all previous models on the RAVEN dataset while using the weakest inductive bias, and it also identifies and addresses a class imbalance in the dataset with a new variant called AB-RAVEN.
We introduce a new neural architecture for solving visual abstract reasoning tasks inspired by human cognition, specifically by observations that human abstract reasoning often interleaves perceptual and conceptual processing as part of a flexible, iterative, and dynamic cognitive process. Inspired by this principle, our architecture models visual abstract reasoning as an iterative, self-contrasting learning process that pursues consistency between perceptual and conceptual processing of visual stimuli. We explain how this new Contrastive Perceptual-Conceptual Network (CPCNet) works using matrix reasoning problems in the style of the well-known Raven's Progressive Matrices intelligence test. Experiments on the machine learning dataset RAVEN show that CPCNet achieves higher accuracy than all previously published models while also using the weakest inductive bias. We also point out a substantial and previously unremarked class imbalance in the original RAVEN dataset, and we propose a new variant of RAVEN -- AB-RAVEN -- that is more balanced in terms of abstract concepts.