Variational Inference for Logical Inference

TL;DR

This paper introduces a variational inference approach within Functional Distributional Semantics to perform logical inference from text, improving computational efficiency and semantic similarity evaluation.

Contribution

It presents a novel method for logical inference using probabilistic evaluation and variational approximation in semantic models, enhancing tractability and performance.

Findings

Logical inference via conditional probabilities is feasible.

Variational approximation speeds up training and inference.

Promising results on semantic similarity tasks.

Abstract

Functional Distributional Semantics is a framework that aims to learn, from text, semantic representations which can be interpreted in terms of truth. Here we make two contributions to this framework. The first is to show how a type of logical inference can be performed by evaluating conditional probabilities. The second is to make these calculations tractable by means of a variational approximation. This approximation also enables faster convergence during training, allowing us to close the gap with state-of-the-art vector space models when evaluating on semantic similarity. We demonstrate promising performance on two tasks.