On Explaining Visual Captioning with Hybrid Markov Logic Networks

TL;DR

This paper introduces a novel, interpretable explanation framework for image captioning models using Hybrid Markov Logic Networks, which identify influential training examples to clarify how captions are generated.

Contribution

It develops a new hybrid logic-based explanation method that combines symbolic rules with real-valued functions to interpret deep neural captioning models.

Findings

The framework effectively identifies training examples influencing caption generation.

Experiments demonstrate the interpretability of explanations for multiple captioning models.

The approach enables comparison of models based on explainability.

Abstract

Deep Neural Networks (DNNs) have made tremendous progress in multimodal tasks such as image captioning. However, explaining/interpreting how these models integrate visual information, language information and knowledge representation to generate meaningful captions remains a challenging problem. Standard metrics to measure performance typically rely on comparing generated captions with human-written ones that may not provide a user with a deep insights into this integration. In this work, we develop a novel explanation framework that is easily interpretable based on Hybrid Markov Logic Networks (HMLNs) - a language that can combine symbolic rules with real-valued functions - where we hypothesize how relevant examples from the training data could have influenced the generation of the observed caption. To do this, we learn a HMLN distribution over the training instances and infer the shift in distributions over these instances when we condition on the generated sample which allows us to quantify which examples may have been a source of richer information to generate the observed caption. Our experiments on captions generated for several state-of-the-art captioning models using Amazon Mechanical Turk illustrate the interpretability of our explanations, and allow us to compare these models along the dimension of explainability.