Pattern recognition systems implemented using deep neural networks achieve better results than linear models. However, their drawback is the black box property. This property means that one with no experience utilising nonlinear systems may need help understanding the outcome of the decision. Such a solution is unacceptable to the user responsible for the final decision. He must not only believe in the decision but also understand it. Therefore, recognisers must have an architecture that allows interpreters to interpret the findings. The idea of post-hoc explainable classifiers is to design an interpretable classifier parallel to the black box classifier, giving the same decisions as the black box classifier. This paper shows that the explainable classifier completes matching classification decisions with the black box classifier on the MNIST and FashionMNIST databases if Zadeh’s fuzzy logic function forms the classifier and DeconvNet importance gives the truth values. Since the other tested significance measures achieved lower performance than DeconvNet, it is the optimal transformation of the feature values to their true values as inputs to the fuzzy logic function for the databases and recogniser architecture used.
The complexity of democratic public decision-making has augmented with the surge of volumes of information. One response is using digital tools in decision-making processes to enable more flexibility than the conventional yes or no responses. We proposed a solution for a large-scale democratic decision-making process using soft computing and fuzzy logic, based on systemic consensus and action design research. Our integrative fuzzy decision-making process was designed to allow the consideration of all the arguments, the deliberation of all the alternatives, and the assessment of each alternative using comparative linguistic expressions. The method was used to resolve a conflict-generating traffic problem in Geuensee, a municipality in the Swiss canton of Luzern. The citizens voted on two dimensions of resistance and support about each of the proposed alternatives. The results were computed using fuzzy membership functions and a fuzzy logic table, evaluated with different computational variants. The output was a ranking of the best options, as assessed by the decision-makers. We found that this method for smart participation of citizens was accepted and generated involvement, leading to an effective outcome for the decision-makers. In the last section, we discuss evaluation and ethical considerations.
Image noise can be viewed as unwanted disturbances in a digital image that should be removed or reduced before further processing and analysis. Impulsive noise, also known as impulse noise, is a very disruptive type of noise, characterized by abrupt variations in brightness in a subset of the image pixels. Impulsive noise commonly occurs during image acquisition and transmission. To mitigate its effects, various impulsive noise reduction methods have been proposed by the image processing community. In contrast to classical filters such as the median filter, most current impulsive noise reduction techniques implement a two-step approach that consists of a noise detection phase to identify noisy pixels and a filtering phase to reduce the amount of noise in the presumably corrupted pixels. The approach presented in this paper is also along this line. To be more precise, we draw on the principles of two state-of-the-art impulsive noise reduction methods, namely the adaptive fuzzy transform based image filter (ATIF) and the improved fuzzy mathematical morphology open-close filter (i-FMMOCS), in order to propose a new method for general impulsive noise reduction.
In this work, we have introduced a new way of speculative reasoning for intelligent systems. In addition, we have illustrated the utility of this way of reasoning in the context of a use case on art genre classification, where explainability and trustworthiness are a matter of major concern. Speculative reasoning is natural for humans and it turns up as a powerful tool, especially in the case of intelligent systems dealing with incomplete, vague, and imprecise information.
Fuzzy Logic, in its fuzzy control incarnation, can be as well seen as an answer to the belated question on how can an algorithm take in account the subtle variations in a complex system – be it the simple and paradigmatic thermostat, the description of a traffic jam or the inner workings of a spaceship. Contrary to the best whodunit, the answer is given at the very start, and the history of the following years of fuzziness is a long demonstration of how, thanks to its explanatory power, a simple idea can be implemented in countless devices, and become the manifest for a technological society in which the logic discourse is based more on the human approach and on embracing the permanent state of flux and uncertainty that is the human experience than on the futile search for absolute truths and endless precision. If the control answer is somehow a given, so many other answers that concern foundations of Fuzzy Logic still beget proper questions. In this paper some of such foundational questions pertaining vagueness, its role in the definition of fuzziness and its many incarnations are set in their historical perspective, and some of the dots outlining the path from a rigid search for truth typical of the end of the nineteenth century to the more nuanced approach that has swept the twentieth century are connected.
