Object Based Attention Through Internal Gating

TL;DR

This paper introduces a neural network model of object-based attention that integrates top-down and recurrent mechanisms, successfully replicating neural phenomena and performing well on both simple and complex visual stimuli.

Contribution

The proposed model uniquely combines top-down and recurrent processes to better emulate biological attention mechanisms in neural networks.

Findings

Replicates neuroscience findings like attention-invariant tuning and inhibition of return.

Performs effectively on simple digit images and complex natural images.

Captures key features of biological object-based attention.

Abstract

Object-based attention is a key component of the visual system, relevant for perception, learning, and memory. Neurons tuned to features of attended objects tend to be more active than those associated with non-attended objects. There is a rich set of models of this phenomenon in computational neuroscience. However, there is currently a divide between models that successfully match physiological data but can only deal with extremely simple problems and models of attention used in computer vision. For example, attention in the brain is known to depend on top-down processing, whereas self-attention in deep learning does not. Here, we propose an artificial neural network model of object-based attention that captures the way in which attention is both top-down and recurrent. Our attention model works well both on simple test stimuli, such as those using images of handwritten digits, and on more complex stimuli, such as natural images drawn from the COCO dataset. We find that our model replicates a range of findings from neuroscience, including attention-invariant tuning, inhibition of return, and attention-mediated scaling of activity. Understanding object based attention is both computationally interesting and a key problem for computational neuroscience.