Affiliation:
1. Paul G. Allen School of Computer Science and Engineering, University of Washington , Seattle, WA 98195 , USA
Abstract
AbstractHuman vision, thought, and planning involve parsing and representing objects and scenes using structured representations based on part-whole hierarchies. Computer vision and machine learning researchers have recently sought to emulate this capability using neural networks, but a generative model formulation has been lacking. Generative models that leverage compositionality, recursion, and part-whole hierarchies are thought to underlie human concept learning and the ability to construct and represent flexible mental concepts. We introduce Recursive Neural Programs (RNPs), a neural generative model that addresses the part-whole hierarchy learning problem by modeling images as hierarchical trees of probabilistic sensory-motor programs. These programs recursively reuse learned sensory-motor primitives to model an image within different spatial reference frames, enabling hierarchical composition of objects from parts and implementing a grammar for images. We show that RNPs can learn part-whole hierarchies for a variety of image datasets, allowing rich compositionality and intuitive parts-based explanations of objects. Our model also suggests a cognitive framework for understanding how human brains can potentially learn and represent concepts in terms of recursively defined primitives and their relations with each other.
Publisher
Oxford University Press (OUP)
Reference52 articles.
1. Human-level concept learning through probabilistic program induction;Lake;Science,2015
2. Building machines that learn and think like people;Lake;Behav Brain Sci,2017
3. Neurocompositional computing: from the central paradox of cognition to a new generation of AI systems;Smolensky;AI Mag,2022
4. A learning algorithm for Boltzmann machines;Ackley;Cogn Sci,1985
5. The Helmholtz machine;Dayan;Neural Comput,1995