Special Session on Advances in Deep Fuzzy Systems

Uzay Kaymak u.kaymak@ieee.org
Eindhoven University of Technology, The Netherlands
João Sousa jmsousa@tecnico.ulisboa.pt
IDMEC, instituto Superior Técnic, Universidade de Lisboa, Portugal
Alexander Gegov alexander.gegov@port.ac.uk
University of Portsmouth, United Kingdom

Programme Description

Deep learning has gained significant attention within the computational intelligence community over the recent years. Its success has been mainly due to the increased capability of modern computers to collect, store and process large volumes of data. This has led to a substantial increase in the effectiveness and efficiency of data management. As a result, it has become possible to achieve high accuracy within a short time frame for some benchmark learning tasks such as object classification and image recognition. The most common implementation of deep learning has been through neural networks due to the ability of their layers to perform multiple functional composition as part of a multistage learning process. However, despite the significant advances in deep learning, there are also limitations. Effectiveness is usually adversely affected when data are not well defined due to inherent noise, uncertainty, ambiguity, vagueness and incompleteness. This has an adverse impact on efficiency due to the necessity to refine the data by additional collection, analysis and cleaning. The reduced effectiveness and efficiency undermine the ability of deep learning to address some real-life tasks that are safety critical or time critical systems. Besides this, deep leaning has been used mainly in a passive manner for the purpose of observing the environment, but it almost has not been used in an active manner for the purpose of changing the environment. Finally, deep learning models often have poor transparency which makes them difficult for understanding, explanation, and interpretation by non-technical users.

Deep fuzzy systems could address some of these problems and limitations. Deep learning can also be used for developing fuzzy systems that can solve more complicated tasks in more dynamic environments. DFS have been around in different forms and under different names such as hierarchical fuzzy systems and networked fuzzy systems. DFS are well suited for performing multiple functional composition at both crisp and linguistic level. Moreover, they have the potential of handling effectively and efficiently data that are not well defined, due to the ability of fuzzy logic to deal successfully with different types of uncertainty. Also, DFS can be used in both passive and active manner with regards to the environment due to their generic structure. Finally, these systems have a high level of transparency due to the ability of fuzzy rules to capture well the interactions between input and output variables.

Scope

This special session aims to bring together contributions on recent advances in deep fuzzy systems (DFS), both from a theoretical and a practical perspective. As such, the session will provide a good outlook on the state-of-the-art in this growing field for the fuzzy systems community. The topics covered, include but are not limited to:

Theoretical methods

  • Hierarchical Fuzzy Systems
  • Networked Fuzzy Systems
  • Chained Fuzzy Systems
  • Multistage Fuzzy Systems
  • Deep learning for Fuzzy Systems

Application areas

  • Object Classification
  • Image Recognition
  • Systems Control
  • Fault Detection
  • Decision Making

Case studies

  • Transport
  • Robotics
  • Business
  • Environment
  • Healthcare
  • Security
  • Energy

Important Dates

Paper submission January 31, 2022
Acceptance/rejection notification April 26, 2022
Camera-ready paper submission May 23, 2022
Conference Dates July 18-23, 2022

Programme Committee

Witold Pedrycz University of Alberta, Canada
Vladik Kreinovich University of Texas, USA
Scott Dick University of Alberta, Canada
Derek Anderson University of Missouri, USA
Uzay Kaymak Eindhoven University of Technology, Netherlands
Joao Sousa University of Lisbon, Portugal
Alexander Gegov University of Portsmouth, UK