Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Pruning Literals for Highly Efficient Explainability at Word Level

Rohan Kumar Yadav, Bimal Bhattarai, Abhik Jana, Lei Jiao, Seid Muhie Yimam

TL;DR

This paper designs a post-hoc pruning of clauses that eliminate the randomly placed literals in the clause thereby making the model more efficiently interpretable than the vanilla Tsetlin Machine.

Abstract

Designing an explainable model becomes crucial now for Natural Language Processing(NLP) since most of the state-of-the-art machine learning models provide a limited explanation for the prediction. In the spectrum of an explainable model, Tsetlin Machine(TM) is promising because of its capability of providing word-level explanation using proposition logic. However, concern rises over the elaborated combination of literals (propositional logic) in the clause that makes the model difficult for humans to comprehend, despite having a transparent learning process. In this paper, we design a post-hoc pruning of clauses that eliminate the randomly placed literals in the clause thereby making the model more efficiently interpretable than the vanilla TM. Experiments on the publicly available YELP-HAT Dataset demonstrate that the proposed pruned TM's attention map aligns more with the human attention map than the vanilla TM's attention map. In addition, the pairwise similarity measure also surpasses the attention map-based neural network models. In terms of accuracy, the proposed pruning method does not degrade the accuracy significantly but rather enhances the performance up to 4% to 9% in some test data.

Pruning Literals for Highly Efficient Explainability at Word Level

TL;DR

This paper designs a post-hoc pruning of clauses that eliminate the randomly placed literals in the clause thereby making the model more efficiently interpretable than the vanilla Tsetlin Machine.

Abstract

Designing an explainable model becomes crucial now for Natural Language Processing(NLP) since most of the state-of-the-art machine learning models provide a limited explanation for the prediction. In the spectrum of an explainable model, Tsetlin Machine(TM) is promising because of its capability of providing word-level explanation using proposition logic. However, concern rises over the elaborated combination of literals (propositional logic) in the clause that makes the model difficult for humans to comprehend, despite having a transparent learning process. In this paper, we design a post-hoc pruning of clauses that eliminate the randomly placed literals in the clause thereby making the model more efficiently interpretable than the vanilla TM. Experiments on the publicly available YELP-HAT Dataset demonstrate that the proposed pruned TM's attention map aligns more with the human attention map than the vanilla TM's attention map. In addition, the pairwise similarity measure also surpasses the attention map-based neural network models. In terms of accuracy, the proposed pruning method does not degrade the accuracy significantly but rather enhances the performance up to 4% to 9% in some test data.

Paper Structure

This paper contains 12 sections, 9 equations, 3 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Proposed Methodology
  4. Tsetlin Machine: Architecture
  5. Pruning Clauses in Tsetlin Machine
  6. Experiments and Results
  7. Explainability
  8. Pairwise Similarity of Attention Map
  9. Performance based on Accuracy
  10. Impact on Explainable Propositional Logic
  11. Conclusion
  12. Limitations

Figures (3)

  • Figure 1: Example of important and non-important literals for given contexts.
  • Figure 2: Representation of TA states for a given context.
  • Figure 3: Case study for three samples and the corresponding clauses with propositional logic before and after pruning. Literals highlighters in yellow color represent that they are removed after the pruning.