Flexible Compositional Learning of Structured Visual Concepts

by   Yanli Zhou, et al.

Humans are highly efficient learners, with the ability to grasp the meaning of a new concept from just a few examples. Unlike popular computer vision systems, humans can flexibly leverage the compositional structure of the visual world, understanding new concepts as combinations of existing concepts. In the current paper, we study how people learn different types of visual compositions, using abstract visual forms with rich relational structure. We find that people can make meaningful compositional generalizations from just a few examples in a variety of scenarios, and we develop a Bayesian program induction model that provides a close fit to the behavioral data. Unlike past work examining special cases of compositionality, our work shows how a single computational approach can account for many distinct types of compositional generalization.


page 1

page 2

page 3

page 4


Learning abstract structure for drawing by efficient motor program induction

Humans flexibly solve new problems that differ qualitatively from those ...

A System for Image Understanding using Sensemaking and Narrative

Sensemaking and narrative are two inherently interconnected concepts abo...

People infer recursive visual concepts from just a few examples

Machine learning has made major advances in categorizing objects in imag...

Numerical Sequence Prediction using Bayesian Concept Learning

When people learn mathematical patterns or sequences, they are able to i...

Radically Compositional Cognitive Concepts

Despite ample evidence that our concepts, our cognitive architecture, an...

Do Vision-Language Pretrained Models Learn Primitive Concepts?

Vision-language pretrained models have achieved impressive performance o...

Human few-shot learning of compositional instructions

People learn in fast and flexible ways that have not been emulated by ma...