Neural Query Language: A Knowledge Base Query Language for Tensorflow

TL;DR

This paper introduces Neural Query Language (NQL), a differentiable framework enabling seamless integration of large knowledge bases into neural models for AI tasks, supporting soft querying, rule incorporation, and template learning.

Contribution

It presents a novel differentiable query language that facilitates neural access to large knowledge bases within gradient-based learning systems.

Findings

Works efficiently with KBs containing millions of tuples.

Supports neural models in adjusting fact confidences.

Enables incorporation of prior knowledge and learning from text.

Abstract

Large knowledge bases (KBs) are useful for many AI tasks, but are difficult to integrate into modern gradient-based learning systems. Here we describe a framework for accessing soft symbolic database using only differentiable operators. For example, this framework makes it easy to conveniently write neural models that adjust confidences associated with facts in a soft KB; incorporate prior knowledge in the form of hand-coded KB access rules; or learn to instantiate query templates using information extracted from text. NQL can work well with KBs with millions of tuples and hundreds of thousands of entities on a single GPU.