Softened Symbol Grounding for Neuro-symbolic Systems
This addresses the fundamental challenge of integrating neural and symbolic components in AI, offering a more effective and efficient approach for neuro-symbolic learning.
The paper tackles the problem of symbol grounding in neuro-symbolic systems by introducing a softened symbol grounding process that bridges neural network training and symbolic constraint solving, resulting in a framework that successfully solves problems beyond existing methods in three representative tasks.
Neuro-symbolic learning generally consists of two separated worlds, i.e., neural network training and symbolic constraint solving, whose success hinges on symbol grounding, a fundamental problem in AI. This paper presents a novel, softened symbol grounding process, bridging the gap between the two worlds, and resulting in an effective and efficient neuro-symbolic learning framework. Technically, the framework features (1) modeling of symbol solution states as a Boltzmann distribution, which avoids expensive state searching and facilitates mutually beneficial interactions between network training and symbolic reasoning;(2) a new MCMC technique leveraging projection and SMT solvers, which efficiently samples from disconnected symbol solution spaces; (3) an annealing mechanism that can escape from %being trapped into sub-optimal symbol groundings. Experiments with three representative neuro symbolic learning tasks demonstrate that, owining to its superior symbol grounding capability, our framework successfully solves problems well beyond the frontier of the existing proposals.