Volltext-Downloads (blau) und Frontdoor-Views (grau)

Squashing activation functions in benchmark tests

  • Over the past few years, deep neural networks have shown excellent results in multiple tasks, however, there is still an increasing need to address the problem of interpretability to improve model transparency, performance, and safety. Achieving eXplainable Artificial Intelligence (XAI) by combining neural networks with continuous logic and multi-criteria decision-making tools is one of the most promising ways to approach this problem: by this combination, the black-box nature of neural models can be reduced. The continuous logic-based neural model uses so-called Squashing activation functions, a parametric family of functions that satisfy natural invariance requirements and contain rectified linear units as a particular case. This work demonstrates the first benchmark tests that measure the performance of Squashing functions in neural networks. Three experiments were carried out to examine their usability and a comparison with the most popular activation functions was made for five different network types. The performance was determined by measuring the accuracy, loss, and time per epoch. These experiments and the conducted benchmarks have proven that the use of Squashing functions is possible and similar in performance to conventional activation functions. Moreover, a further experiment was conducted by implementing nilpotent logical gates to demonstrate how simple classification tasks can be solved successfully and with high performance. The results indicate that due to the embedded nilpotent logical operators and the differentiability of the Squashing function, it is possible to solve classification problems, where other commonly used activation functions fail.

Export metadata

Additional Services

Share in Twitter Search Google Scholar

Statistics

frontdoor_oas
Metadaten
Author:D. Zeltner, B. Schmid, G. Csiszar, O. Csiszar
URL:https://arxiv.org/abs/2010.08760
Parent Title (English):Knowledge-Based Systems
Subtitle (English):towards explainable Artificial Intelligence using continuous-valued logic
Publisher:arXiv
Document Type:Article
Language:English
Year of Completion:2020
Release Date:2021/01/29
Pagenumber:19
Institutes:Fakultäten / Grundlagen
Open Access:nur im Hochschulnetz
Relevance:Keine peer reviewed Publikation (Wissenschaftlicher Artikel und Aufsatz, Proceeding, Artikel in Tagungsband)
Licence (German):License LogoVeröffentlichungsvertrag ohne Print-on-Demand