DeepAI AI Chat
Log In Sign Up

Neural-Symbolic Integration: A Compositional Perspective

by   Efthymia Tsamoura, et al.

Despite significant progress in the development of neural-symbolic frameworks, the question of how to integrate a neural and a symbolic system in a compositional manner remains open. Our work seeks to fill this gap by treating these two systems as black boxes to be integrated as modules into a single architecture, without making assumptions on their internal structure and semantics. Instead, we expect only that each module exposes certain methods for accessing the functions that the module implements: the symbolic module exposes a deduction method for computing the function's output on a given input, and an abduction method for computing the function's inputs for a given output; the neural module exposes a deduction method for computing the function's output on a given input, and an induction method for updating the function given input-output training instances. We are, then, able to show that a symbolic module – with any choice for syntax and semantics, as long as the deduction and abduction methods are exposed – can be cleanly integrated with a neural module, and facilitate the latter's efficient training, achieving empirical performance that exceeds that of previous work.


page 1

page 2

page 3

page 4


Neuro-Symbolic Program Synthesis

Recent years have seen the proposal of a number of neural architectures ...

Compositional Generalization via Neural-Symbolic Stack Machines

Despite achieving tremendous success, existing deep learning models have...

Harmonic Grammar in a DisCo Model of Meaning

The model of cognition developed in (Smolensky and Legendre, 2006) seeks...

A Context-Dependent Gated Module for Incorporating Symbolic Semantics into Event Coreference Resolution

Event coreference resolution is an important research problem with many ...

The Refinement Calculus of Reactive Systems

The Refinement Calculus of Reactive Systems (RCRS) is a compositional fo...

Symbolic Synthesis of Neural Networks

Neural networks adapt very well to distributed and continuous representa...

Meta Module Network for Compositional Visual Reasoning

There are two main lines of research on visual reasoning: neural module ...