Object-Centric Learning with Slot Attention
This addresses the challenge of enabling efficient abstract reasoning from perceptual features in AI, representing a novel method for a known bottleneck in object-centric learning.
The paper tackles the problem of learning object-centric representations from complex scenes by introducing the Slot Attention module, which produces exchangeable slots that bind to objects through competitive attention, and demonstrates improved generalization to unseen compositions in unsupervised object discovery and supervised property prediction tasks.
Learning object-centric representations of complex scenes is a promising step towards enabling efficient abstract reasoning from low-level perceptual features. Yet, most deep learning approaches learn distributed representations that do not capture the compositional properties of natural scenes. In this paper, we present the Slot Attention module, an architectural component that interfaces with perceptual representations such as the output of a convolutional neural network and produces a set of task-dependent abstract representations which we call slots. These slots are exchangeable and can bind to any object in the input by specializing through a competitive procedure over multiple rounds of attention. We empirically demonstrate that Slot Attention can extract object-centric representations that enable generalization to unseen compositions when trained on unsupervised object discovery and supervised property prediction tasks.