IPMU2026

Special Sessions

The rapid growth of information and communication technologies generates vast amounts of data every day. The complexity and uncertainty of real-world data bring both challenges and opportunities for researchers and companies. 

Researchers and companies study and create algorithms and tools tailored to process and analyse the intricate nature of real-world data. These efforts aim to improve downstream tasks in data, text, and social network mining, such as clustering, sentiment analysis, topic modelling,document classification, text summarisation, speech recognition, community detection, influence analysis, anomaly detection and more.

In the contemporary world, data, text, and social network mining solutions are integrated into nearly every domain. They transcend boundaries, finding applications in realms as diverse as politics, sports, healthcare, and beyond. The inherent imprecision of real-world data adds additional complexity, urging researchers and developers to navigate the intricacies of real-data nuances, ensuring that these solutions are robust and attuned to the subtleties of real-world data across various domains.

Topics of interest include, but are not limited to, the following topics:

• – Flexible data, text, web mining and big data mining
• – Data text and web mining approaches under uncertainty.
• – Stream data mining and temporal data series
• – Imprecision, uncertainty, and vagueness in data mining
• – Data pre- and post-processing in data mining
• – Parallel and distributed data mining algorithms
• – Information summarisation and visualisation
• – Human-machine interaction for data access
• – Semantic models to represent input data and extracted knowledge in a data mining process
• – Applications of data mining techniques: health, nutrition, tourism, biological and industrial processes, customer profiles, anomaly detection, emergency management, event detection, etc.
• – Multilinguality
• – Information retrieval, search and question-answering
• – Multimodality and data fusion strategies
• – Interpretability and analysis of models in data mining
• – Challenges and advances in NLP for social media: Fake news , hate speech, sexism detection, and beyond.
• – Community detection, influence analysis and anomaly detection

Aggregation theory has become a central branch of fuzzy set theory. Traditionally, the field has focused on aggregation functions as tools for combining degrees of truth or membership values, typically defined within the unit interval. Consequently, much of the developed theory has concentrated on functions over this unit interval or, more generally, over closed and bounded real intervals.

In recent decades, however, researchers across a wide range of disciplines have extended aggregation processes to richer and more complex structures. A non-exhaustive list of examples is the following: compositional data in geochemistry, directional data in biology, functional data in functional data analysis, image data in computer vision, multivariate data in statistics, ranking data in social choice theory, and string data in computer science.

This special session builds on the strong tradition of aggregation theory sessions at past IPMU conferences by providing a forum for exploring aggregation beyond the conventional restriction to bounded real intervals. We invite both applied and theoretical contributions, especially those addressing unconventional data types or advancing the integration of aggregation functions into data summarization and other statistical techniques.

The proposed special session focuses on the development and application of statistical methods, multicriteria decision-making techniques, and artificial intelligence approaches to understand and model individual and collective decision-making in political, institutional, and social contexts. The increasing availability of open data, together with recent advances in probabilistic modeling, fuzzy logic, machine learning, and network analysis, enables the exploration of interactions between individual preferences and aggregate outcomes, as well as the measurement of consensus, disagreement, and strategic behavior in complex environments.

This session seeks contributions that advance our understanding of individual and collective decision processes by introducing innovative frameworks, statistical techniques, or computational tools for handling uncertainty, preference aggregation, and complex voting scenarios. Both methodological and applied studies are encouraged. 

Topics of interest include but are not limited to:

• – Statistical models for voting behavior and collective decisions (Bayesian, spatiotemporal, multilevel models).
• – Ecological inference and small-area estimation.
• – Fuzzy, intuitionistic, and other non-classical approaches to capture consensus and disagreement.
• – Multicriteria decision-making methods under uncertainty (e.g., PROMETHEE II, fuzzy MCDM).
• – Machine learning algorithms for election prediction, opinion analysis, and preference aggregation.
• – Simulation and experimental frameworks to study strategic voting or hypothetical scenarios.
• – Advanced visualization of electoral and social data, network analysis of influence and opinion formation.
• – Ethical considerations, transparency, and reproducibility in decision data analysis.

This session is designed to foster an interdisciplinary exchange among researchers from statistics, political science, economics, data science, and decision theory, promoting discussion on new methodologies and applications in political, social, and economic contexts.

The rapid advancement of artificial intelligence has produced increasingly powerful predictive models, yet their deployment in critical decision-making contexts—such as healthcare, finance, law, and public policy—remains hindered by two fundamental challenges: lack of interpretability and uncertainty quantification.

This special session aims to bridge the gap between Explainable AI (XAI) and decision-making under uncertainty, bringing together methodological innovations and practical applications that address the dual need for transparent and robust intelligent systems.

Key themes include:

• – XAI methodologies for uncertain and imprecise data: integrating fuzzy sets, rough sets, belief functions, and possibility theory with explainability frameworks
• – Interpretable machine learning models in the presence of uncertainty: rule-based systems, transparent neural architectures, and symbolic reasoning
• – Uncertainty-aware decision support systems: propagation of epistemic and aleatory uncertainty through AI pipelines
• – Trustworthy AI under imprecision: combining XAI techniques with imprecise probabilities, evidential reasoning, and multi-criteria decision analysis
• – Human-AI interaction in uncertain environments: designing explainable outputs that support human decision-makers
• – Applications: medical diagnosis, financial risk assessment, climate modeling, smart cities, and critical infrastructure management

The session welcomes theoretical contributions, algorithmic developments, and real-world case studies that demonstrate how explainability and uncertainty management can be effectively integrated to produce AI systems that are not only accurate but also interpretable, reliable, and aligned with human values.

By fostering dialogue between XAI researchers and uncertainty quantification experts, this session aims to advance the state-of-the-art in building decision-support tools that practitioners can trust and regulators can validate.

For more than a decade, fuzzy implication functions have been one of the main research lines within the fuzzy logic community. These logical connectives generalize the classical two-valued implication to the infinite-valued setting. In addition to modelling fuzzy conditionals, they are also used to perform backward and forward inferences in various fuzzy rule-based systems. Moreover, they have proved useful not only in fuzzy control and approximate reasoning, but also in many other fields such as multi-valued logic, image processing, data mining, computing with words, and rough sets, among others. 

Due to this wide range of applications, fuzzy implication functions have attracted the attention of many researchers from both theoretical and applied perspectives. Indeed, the theoretical viewpoint focuses on problems whose solutions provide important insights for practical applications. Therefore, this special session seeks to bring together researchers interested in recent advances in the theory and applications of fuzzy implication functions, including, among others, their characterizations, representations, generalizations, and relationships with fuzzy negations, triangular norms, uninorms, and other fuzzy logic connectives.

Keywords and Related Topics: 

• – Fuzzy implication functions
• – Characterizations
• – New construction methods
• – Relationships with other fuzzy logic connectives
• – Novel structural relationships
• – Approximate reasoning
• – Applications of fuzzy implication functions: image processing, data mining, etc.

Mastering complexity and uncertainty in complex IT systems remains challenging. This special track invites researchers to explore the intersection of Fuzzy Logic (FL), which models epistemic uncertainty, and Quantum Computing (QC), which confronts the inherent ontological uncertainty of reality itself.

The current NISQ era limits performance due to errors. We believe that FL, with its unique ability to handle imprecision, offers a novel approach to improving these systems. We encourage submissions that build bridges between the theoretical foundations, methodologies, and applications of FL and QC.

We understand artificial intelligence as a technology that enables machines to perform tasks requiring human intelligence, such as decision making, learning, reasoning, natural language processing, problem solving, and so on. We argue that fuzzy modeling methods can make a significant contribution to solving these problems due to their ability to account for uncertainty—a phenomenon inherent in linguistic semantics and human reasoning. 

In this context, we plan to approach XAI based on understanding the meaning of interrelated steps, starting with the formulation of the problem to be solved (including natural language) and continuing through modeling, computation, and analysis of the results. In this problem-solving chain, we focus on modeling as a fuzzy approximation in the general sense, formulated in terms of various theories, such as fuzzy natural logic, fuzzy concept analysis, and fuzzy transforms. 

With this in mind, we plan to discuss the use of neural networks as basic computational models (including, but not limited to, fuzzy models) and, consequently, the practical aspects of learning model parameters. This special section will provide a forum for discussing new directions regarding the role of fuzzy modeling methods in artificial intelligence theory and its applications. It will cover (but is not limited to) the following topics:

• – Logical theory of the meaning of natural language expressions, including those with generalized quantifiers, having the form of logical syllogisms
• – Data representation using general tools of fuzzy modeling, including: fuzzy natural logic, fuzzy transforms as a method of the compressed data representation, fuzzy concept analys
• – Extended fuzzy concept analys
• – Logical structures of opposition with generalized quantifiers and their applications
• – Large language models and their processing

Intelligent control systems combine fuzzy logic with machine learning, optimization, and advanced control methods. Fuzzy logic remains essential thanks to its ability to encode expert knowledge, manage uncertainty, and enable practical deployments from edge devices to cloud platforms. This special session invites theoretical and applied contributions on fuzzy modelling/identification and fuzzy control, as well as their synergies with complementary techniques (neuro-fuzzy, reinforcement learning, MPC, estimation, etc.). We emphasize robustness, interpretability, verification, and real-time execution for complex systems in industrial and service domains.

Topics of Interest (include but are not limited to)

• – Fuzzy control: Mamdani, Takagi–Sugeno, adaptive/supervisory control, gain scheduling, anti-windup.
• – Fuzzy MPC: centralized, distributed/coalitional, robust and stochastic; multi-agent coordination.
• – Modelling and identification: ANFIS and neuro-fuzzy approaches, type-2 fuzzy (interval and general), inverse identification, hybrid physics-informed + data-driven models, fuzzy clustering.
• – High-dimensional systems: feature selection and reduction, sparsity, rule compression, regularization and structure learning.
• – Learning and adaptation: online training, transfer/meta-learning, federated and edge learning; RL with fuzzy rules; calibration and uncertainty quantification.
• – Observers and estimation: fuzzy observers, filtering, set-membership approaches, integration with Bayesian filters.
• – Optimization and decision-making: multi-objective optimization, planning and scheduling, decision under uncertainty, explainability and rule auditing.
• – Verification, validation, and certification: formal property analysis, scenario-based testing, V&V for real-time systems.
• – Reliability, safety, and cyber-resilience: fault diagnosis, anomaly detection, fault tolerance, secure control, and privacy preservation.
• – Implementation and deployment: edge/embedded realizations, FPGA/SoC, HW/SW co-design, interoperability in IIoT/CPS.
• – Data and quality: robust preprocessing, data/Concept drift handling, virtual sensors.
• – Applications: process industry, robotics, automotive, energy and microgrids, buildings and smart cities, water, healthcare, agri-food, logistics.
• – (Included) Digital Twins: architectures and physical–virtual synchronization, co-simulation, online updating of rules and membership functions, lifecycle data quality management.

In many real-life problems one needs to process information which is affected by some type of uncertainty. Fuzzy sets were introduced by L. Zadeh with the aim of developing a mathematical tool for modeling the vagueness and, thus, to overcome the limitations of classical theory when modelling the aforesaid uncertainty arising in practical problems. Since then, numerous studies have been carried out on different theoretical aspects of fuzzy sets theory (operations, representation techniques, aggregation methods, fuzzy measures, fuzzy relations, etc), which has consequently experienced growing interest and come to play a crucial role in many fields of science today. 

All these developments have been closely linked to an increasing number of applications to many different topics, image processing, machine learning, decision making, pattern recognition, medical diagnosis, social sciences and robotics, just to mention a few of them.

The goal of this special session is to provide a forum where researchers can exchange ideas and approaches, and share the latest developments focused on the study of the concept of fuzzy set and its generalizations, as well as associated concepts such as fuzzy measures or fuzzy relations, from theoretical and applied viewpoints. Special attention is paid to aggregation information methods. 

For reference purpose, below is a non-exhaustive list of topics.

Theoretical aspects on:

• – Fuzzy sets and extensions (Intuitionistic, Picture, Pythagorean, Spherical, 2-type, Interval-valued, etc.)
• – Fuzzy measures
• – Fuzzy relations
• – Fuzzy Inference Systems
• – Fuzzy relations and generalized metric spaces: duality relationship
• – Aggregation methods in a broad sense: generalizations of the idea of aggregation function, aggregation functions for extensions of fuzzy sets, aggregation of fuzzy relations, aggregation of structures, etc.

Practical aspects on:

• – Approximate reasoning
• – Image processing
• – Model identification and parametrization
• – Diagnostics and prognostics
• – Data mining
• – Decision-Making
• – Clustering under uncertainty
• – Multi-agent systems
• – Robotics
• – Machine Learning
• – Pattern Recognition
• – Swarm Intelligence
• – Business and Financial Analysis

In today's society, many problems are hard to solve because they have several complex characteristics (beyond their size). These characteristics include imprecision, uncertainty, different types and sources of information, dynamism, among others.

One way to address such issues is through the use of cooperative automated systems (CAS) that integrate multiple independent strategies (IS). These IS solve the same problem but employ different criteria and algorithms. By combining the outputs of the IS, and thus considering simultaneously a wide range of aspects of the original problem, a decision or solution can be obtained. The use of CAS is expected to result in more robust and less imprecise final decisions, as they will be based on a more comprehensive set of evidences.

It's essential to recognize the versatility and potential of CAS for problem solving. For instance, when multiple multicriteria decision-making methods are employed simultaneously, and their corresponding rankings are combined using aggregation techniques, a more comprehensive and robust outcome can be achieved. Similarly, the use of multiple metaheuristics methods, followed by the application of decision rules taking into account unmodeled aspects during the solving stage, can lead to a more informed and effective solution. Furthermore, the output type of a CAS can differ from those produced by IS. Departing from IS returning real values as output, a CAS may return an interval, a fuzzy number, etc. This flexibility is a significant advantage, as it allows decision-makers to consider a broader spectrum of potential outcomes and make more informed decisions. 

Several questions are open in the design and application of these cooperative methods. For example:

• a) What IS must be defined to implement a CAS?
• b) What type of information (numbers, intervals, labels, etc.) is managed at each level?
• c) How the output of the IS should be combined? What happened if the information is heterogenous?
• d) How can the reliability and relevance of the IS could be characterized and considered?

The aim of this section is to promote the discussion of both the up-to-date theoretical research in these aspects, as well as CAS applications in fields like Economy, Transportation, Health, Education, etc.. The framework of CAS is ideal to explore the many topics arising at the IPMU conference including (but not limited) management/representation/measuring imprecision and uncertainty in the input/output information, aggregation mechanisms for the combination of the outputs, exploring ways to measure reliability or confidence for the individual components, and so on.

The search of new information fusion techniques under uncertainty is currently a hot topic in almost every research field, from image processing, classification, data stream clustering, brain computer interfaces, decision making to deep learning and adaptive neuro fuzzy inference systems. This interest has led to new analysis of the notion of aggregation function and the introduction of new concepts that go beyond usual aggregation functions, either by considering more general definitions (e.g., considering weaker forms of monotonicity), or by extending them to other frameworks different from that of the unit interval (e.g., intervals, lattices). The aim of this section is to promote the discussion of the up-to-date theoretical research in the topic, as well as their applications, in total connection with the interests of IPMU 2026, related to the theoretical and applied subjects covered by conference.

Interval uncertainty is closely related to fuzzy techniques: indeed, if we want to know how the fuzzy uncertainty of the inputs propagates through the data processing algorithm, then the usual Zadeh's extension principle is equivalent to processing alpha-cuts (intervals) for each level alpha. 

This relation between intervals and fuzzy computations is well known, but often, fuzzy researchers are unaware of the latest most efficient interval techniques and thus use outdated less efficient methods. One of the objectives of the proposed session is to help fuzzy community by explaining the latest interval techniques and to help interval community to better understand the related interval computation problems. 

Yet another relation between interval and fuzzy techniques is that the traditional fuzzy techniques implicitly assume that experts can describe their degree of certainty in different statements by an exact number. In reality, it is more reasonable to expect experts to provide only a rage (interval) of possible values – leading to interval-valued fuzzy techniques that, in effect, combine both types of uncertainty.

The Special Session "Knowledge representation and modelling” is focused on theoretical and applied tools for representing and modelling information. In particular, our interest is in the direction of recent techniques for dealing with uncertainty. In this sense, Formal Concept Analysis, Logic Programming and Fuzzy Relation Equations, together with their fuzzy extensions, arise as reliable tools for dealing with knowledge obtained from databases that are uncertain in some way, such as incomplete, imprecise, ambiguous and graded.

This special session aims to bring together contributions that explore how uncertainty and imprecision can be effectively modeled, quantified, and managed to support reliable and responsible decision-making in complex, data-rich environments. 

The capacity to reason and act under imperfect, incomplete, or ambiguous information is fundamental across diverse areas of knowledge, from data science and engineering to life sciences and the social domain. 

The session welcomes both theoretical and applied works, including (but not limited to) the following perspectives:

• – Artificial Intelligence and Machine Learning: integration of uncertainty estimation, interpretability, and robustness to enhance model reliability.
• – Decision Support and Risk Analysis: frameworks and computational tools for decision-making under uncertainty in dynamic or high-stakes contexts.
• – Mathematical and Conceptual Foundations: new measures, representations, or theoretical links between uncertainty, imprecision, and reliability.
• – Applications: interdisciplinary studies where uncertainty modeling is key — including environmental, biomedical, industrial, and socio-technical systems.

By fostering cross-domain dialogue, this session seeks to advance methods and perspectives that connect uncertainty modeling with trustworthy and effective decision processes across science, technology, and society.

Multi-valued logics (MVLs) extend classical logic by admitting a spectrum of truth degrees, offering flexible tools to reason under ambiguity, inconsistency, and incomplete knowledge. Their semantics, ranging from fuzzy truth values to ordinal plausibility measures, make MVLs a natural setting to explore links between semantics, algebraic structures, and logic. Increasingly, MVLs are also combined with probabilistic methods, where random variables model the distribution of truth degrees, bridging logical reasoning and statistical inference.

This special session aims to include papers related to MVLs that either present significant advances in the foundations or demonstrate potential applications in real-world problems. Contributions highlighting the interplay between MVLs, imprecise probability theories, and fuzzy sets — especially in contexts such as knowledge representation and decision making — are especially welcome.

This invited session focuses on Multicriteria Decision Aiding (MCDA), a set of methods and tools designed to support complex decision-making processes involving multiple, often conflicting criteria. The session will cover theoretical foundations, practical applications, and innovative methodologies for evaluating and comparing alternatives. Special emphasis will be placed on the interplay between MCDA and Artificial Intelligence (AI), highlighting how machine learning, knowledge-based representations, and intelligent optimization can enhance MCDA approaches. Moreover, the session will explore aspects of interpretability and explainability, addressing how AI-augmented MCDA can provide transparent, understandable, and justifiable decision support in uncertain and complex environments.

Recent advances in robotics and autonomous systems have brought machines increasingly close to human environments, requiring them to act, decide, and communicate under multiple layers of uncertainty. While traditional research in uncertainty management has focused on data, logic, and deduction / inference, human–machine interaction introduces new challenges — uncertainty becomes embodied, social, and affective. Robots and autonomous systems must not only reason about uncertain sensory inputs or incomplete models of the world, but also interpret and adapt to human emotions, intentions, and unpredictable behaviors.

This special session seeks to explore how uncertainty shapes human–machine understanding, from perception and motion planning to social interaction and affective communication. It aims to bring together AI with robotics researchers and specialists in psychology, neuroscience, and social sciences to study how embodied autonomous systems can sense, reason about, and communicate uncertainty in ways that align with human cognition and emotion. 

By extending traditional uncertainty management into embodied and interactive contexts, this session addresses fundamental questions: How can robots express uncertainty in interpretable ways? How can they reason about human affect or intent when cues are ambiguous or incomplete? How can uncertainty be leveraged to build systems that are not only reliable and adaptive but also socially and ethically aware? 

Through this multidisciplinary exchange, the session will foster new frameworks for representing, reasoning with, and communicating uncertainty in embodied interaction, paving the way toward more transparent, trustworthy, and human-aligned autonomous systems.

Topics of interest include, but are not limited to:

• – Managing perceptual and predictive uncertainty in dynamic human environments
• – Embodied representations of uncertainty in robotic motion and action planning
• – Communicating and adapting to uncertainty in human–robot collaboration
• – Emotional and affective uncertainty in human–machine dialogue and interaction
• – Cognitive architectures for reasoning under uncertainty in social contexts
• – Trust calibration, transparency, and adaptation in uncertainty-aware robotic systems
• – Social and ethical implications of uncertainty management in autonomous systems
• – Evaluation and benchmarking of uncertainty-aware interaction models

This special session focuses primarily on mathematical models for economic and financial applications, under conditions of ambiguity, partial, imprecise, and revisable knowledge. Uncertainty in the quoted applications is managed through non-additive uncertainty measures so as to achieve robustness.

These measures are particularly relevant for the construction of non-linear functionals, which allow for the introduction of behavioral decision-making models, as well as risk measures and models for incomplete or frictional markets. The problem of conditioning reveals also to be crucial for developing dynamic decision-making models. Besides that, the quoted models find a distinguished use to address robust machine learning techniques.

Examples of target applications are:

• – portfolio choice under ambiguity;
• – pricing in frictional markets of the bid-ask spread type;
• – coherent risk measurement;
• – robust reinsurance problems;
• – similarity and divergence measures in fuzzy and/or imprecise environments;
• – strategic decisions in the presence of partial knowledge;
• – robust machine learning techniques in economics and finance.

The increasing complexity of scientific and practical problems, accompanied by the growing diversity of data types, presents significant challenges for researchers and practitioners across disciplines such as engineering, the social sciences, and medicine. Addressing these challenges necessitates the development of advanced methodologies and tools capable of modeling and processing various forms of uncertainty – including randomness, imprecision, and ambiguity – while effectively integrating modern computational techniques and machine learning approaches. This Special Session seeks to bring together theorists and practitioners in the fields of statistical reasoning and data analysis to discuss emerging challenges, exchange innovative ideas, and foster the development of flexible inferential and soft computing methods that enhance understanding and provide robust solutions to complex real-world problems.

Topics of interest include but are not limited to:

• – Analysis of censored or missing data
• – Analysis of fuzzy data
• – Bayesian methods
• – Clustering and classification
• – Data mining
• – Fuzzy random variables
• – Fuzzy regression methods
• – Granular computing
• – Interval data
• – Machine learning
• – Possibility theory
• – Random sets
• – Robust statistics
• – Semi-supervised learning
• – Streaming data
• – Statistical software for imprecise data
• – Time series analysis and forecasting

This session is dedicated to Imprecise Probability Theory, which encompasses a wide range of mathematical models that offer a flexible alternative to classical Probability Theory when the available information is scarce, vague, or incomplete. Among the many models included in this general framework, it is worth highlighting lower previsions, non-additive measures, (2- or n-monotone) capacities, belief functions, possibility measures, and p-boxes, among others. 

The aim of this special session is to gather papers on Imprecise Probabilities that either present significant advances in the theoretical foundations or demonstrate potential applications to real-world problems. Contributions that emphasise the connection between imprecise probability theories and other fields, such as game theory, decision-making, or fuzzy sets, are also welcome.

We propose a special session in connection with the EUSFLAT Working Group on Aggregation Functions for the IPMU 2026 conference. Aggregation functions are fundamental tools in decision-making, information fusion, data analysis, and risk assessment, particularly in contexts involving uncertainty, imprecision, and conflicting information, making this a highly relevant topic for the IPMU community. This special session aims to gather researchers and practitioners presenting cutting-edge theoretical developments and innovative applications of aggregation functions. We encourage submissions that address new challenges and extend the classical theoretical framework of these functions. 

The core topics of interest include, but are not limited to:

• – Generalised monotonicity and axiomatic properties
• – Novel aggregation function families
• – Practical applications and computational aspects

Topics in the fields of machine learning and data mining have attracted considerable attention within the fuzzy set community in recent years. There are several motivations for combining tools and techniques from fuzzy set theory with learning and data mining methods, notably the following: Firstly, learning and adaptivity have become important aspects in fuzzy systems design, where data-driven approaches can complement knowledge-based methods in a reasonable way. Secondly, recent research has shown that fuzzy set theory can contribute to machine learning and data mining in a substantial way, e.g., in dealing with uncertainty in model induction or extracting vague patterns and relationships from data. The general goal of the working group is to promote research in the field of fuzzy machine learning and data mining. Moreover, the working group shall provide a forum for discussions on this topic and a repository for resources on fuzzy data mining, including, e.g., software and benchmark data sets.

This special session is devoted to the most recent developments of Mathematical Fuzzy Logic. Thus, to formal fuzzy logics from a mathematical point of view. We encourage particular emphasis on theoretical advances related to many-valued logics, algebraic semantics, combinatorial aspects, topological and categorical methods, proof theory and game theory, many-valued computation, many-valued logics and finite model theory, many-valued logics and AI. 

This special session is dedicated to Jan Łukasiewicz (1878-1956), an important figure in the area of Mathematical Fuzzy Logic.

A partial list of topics is the following:

• – Algebraic semantics of fuzzy logics
• – First-order fuzzy logics, with applications to model theory and finite model theory
• – Fuzzy modal logics
• – Proof theory for fuzzy logics
• – Combinatorial or topological dualities
• – Computational complexity of many-valued logics
• – Fuzzy logic approaches to (subjective) probability and general uncertainty models
• – Fuzzy logics and automated reasoning
• – Fuzzy logics and AI

This session aims at gathering researchers interested and involved in uncertainty issues in processing and analysis of images, in various applications. Uncertainty (in a broad sense) can pertain to data, to knowledge guiding their analysis, to processing steps and to the analysis results.

At the same time, this session will discuss methods for improving image quality, which is affected by inaccuracies caused by image acquisition or processing. We are talking about eliminating visible defects (noise, blurred edges, etc.) and restoring the original information contained in the image.

The theoretical issues are at the core of many topics. Here the aim of the session would be address them in the specific context of image analysis, and more generally spatial information processing.

This special session is a joint event with the EUSFLAT Working Group on Soft Computing in Image Processing, which will encourage EUSFLAT members to contribute and promote discussions.