Two conflicting goals are often involved in the design of fuzzy rule-based systems: Accuracy maximization and interpretability maximization. A number of approaches have been proposed for finding a fuzzy rule-based system with a good accuracy-interpretability tradeoff. Formulation of the accuracy maximization is usually straightforward in each application area of fuzzy rule-based systems such as...

We study the problem of inducing interpretability in KG embeddings. Specifically, we explore the Universal Schema (Riedel et al., 2013) and propose a method to induce interpretability. There have been many vector space models proposed for the problem, however, most of these methods don’t address the interpretability (semantics) of individual dimensions. In this work, we study this problem and p...

We investigate the modal logic of interpretability over Peano arith metic PA Our main result is an extension of the arithmetical complete ness theorem for the interpretability logic ILM This extension concerns recursively enumerable sets of formulas of interpretability logic rather than single formulas As corollaries we obtain a uniform arithmetical completeness theorem for the interpretability...

articles in general and definite articles in particular can create problems even long after all other aspects of english have been mastered. the present article investigated the learnability problems related to the acquisition of count-mass distinction of english nominals by persian l2 learners. the theoretical underpinning of the study is the interpretability hypothesis (tsimpli & dimitrak...

The aim of this contribution is to point out the importance of interpretability of fuzzy rule-based systems. We try to approach this key property axiomatically, as a starting point, restricting to interpretability of linguistic variables (in Zadeh’s sense). The proposed approach is underlined with an intuitive example. Finally, the benefits of considering interpretability in design and tuning o...

Interpretability is often a major concern in machine learning. Although many authors agree with this statement, interpretability is often tackled with intuitive arguments, distinct (yet related) terms and heuristic quantifications. This short survey aims to clarify the concepts related to interpretability and emphasises the distinction between interpreting models and representations, as well as...

Fuzzy Logic Based Classification System usually comes with two contradictory requirements to the obtained model; namely Accuracy and Interpretability. It is often found that a fuzzy system with highest classification accuracy shows poorer interpretability and vice versa. This is also observed in the hybrid GA-PSO approaches presented in the previous chapter. IGA has reasonable interpretability ...