Universal Topological Regularities of Syntactic Structures: Decoupling Efficiency from Optimization

no code implementations • 31 Jan 2023 • Fermín Moscoso del Prado Martín

Among the two options, efficiency without optimization offers the better explanation for the new pattern.

Trepan Reloaded: A Knowledge-driven Approach to Explaining Artificial Neural Networks

no code implementations • 19 Jun 2019 • Roberto Confalonieri, Tillman Weyde, Tarek R. Besold, Fermín Moscoso del Prado Martín

Whilst a plethora of approaches have been developed for post-hoc explainability, only a few focus on how to use domain knowledge, and how this influences the understandability of global explanations from the users' perspective.

Decision Making Interpretable Machine Learning

Constant conditional entropy and related hypotheses

no code implementations • 27 Apr 2013 • Ramon Ferrer-i-Cancho, Łukasz Dębowski, Fermín Moscoso del Prado Martín

We show that constant entropy rate (CER) and two interpretations for uniform information density (UID), full UID and strong UID, are inconsistent with these laws.