Learning Continuous Semantic Representations of Symbolic Expressions
This addresses the grand challenge of combining symbolic and neural reasoning for AI researchers, though it appears incremental as a step in this direction.
The paper tackles the challenge of learning continuous semantic representations of symbolic expressions by proposing neural equivalence networks, which significantly outperform existing architectures on equivalence checking across diverse algebraic and boolean expressions.
Combining abstract, symbolic reasoning with continuous neural reasoning is a grand challenge of representation learning. As a step in this direction, we propose a new architecture, called neural equivalence networks, for the problem of learning continuous semantic representations of algebraic and logical expressions. These networks are trained to represent semantic equivalence, even of expressions that are syntactically very different. The challenge is that semantic representations must be computed in a syntax-directed manner, because semantics is compositional, but at the same time, small changes in syntax can lead to very large changes in semantics, which can be difficult for continuous neural architectures. We perform an exhaustive evaluation on the task of checking equivalence on a highly diverse class of symbolic algebraic and boolean expression types, showing that our model significantly outperforms existing architectures.