Complex computational systems are often intrinsically unpredictable even for the engineers designing them: understanding their possible evolution over time is typically required, for instance in order to match some application requirements. In this paper, we first survey some of the most promising and effective techniques for the analysis and prediction of complex computational systems, then we focus on SAPERE-coordinated pervasive ecosystems. We identify the most essential elements of the SAPERE coordination model, select two techniques – model checking and data mining –, and derive two possible approaches to the analysis and prediction of the behaviour of SAPERE systems.
The document discusses using ontology to enhance management of operating system services. It proposes an ontology-based architecture to classify and organize various operating system services. The architecture divides services into the kernel and exokernel layers. Individual service ontologies are developed for search, backup, security, and networking and then merged into a universal ontology. This approach aims to improve service development, deployment and management by adding semantics and enabling knowledge sharing and reusability across operating systems.
The document summarizes and compares different methods for face recognition, including Eigenface, Line Edge Map (LEM), and other techniques. It provides descriptions of how each technique works, such as using eigenvectors to extract features for Eigenface. Experimental results show LEM achieves better accuracy than Eigenfaces for variations in lighting and size. While Eigenfaces struggles with size changes, LEM maintains high accuracy for different conditions. The document recommends future work combining techniques to maximize recognition accuracy.
Molecules of Knowledge: Self-Organisation in Knowledge-Intensive EnvironmentsAndrea Omicini
We propose a novel self-organising knowledge-oriented model based on biochemical tuple spaces, called Molecules of Knowledge (MoK). We introduce MoK basic entities, define its computational model, and discuss its mapping on the TuCSoN coordination model for its implementation.
[IDC 2012, Calabria, Italy, 24/9/2012
Autonomy, Interaction & Complexity in Computational Systems. Preliminary NotesAndrea Omicini
The issues of autonomy do not end with autonomous components. Autonomous systems are complex self-organising systems where interaction plays an essential role. Here we discuss some early thoughts on autonomy, interaction, and complexity in artificial systems, and suggest that nature-inspired coordination models should work as the main sources of technologies.
Nature-inspired Coordination for Complex Distributed SystemsAndrea Omicini
Originating from closed parallel systems, coordination models and technologies gained in expressive power so to deal with open distributed systems. In particular, nature-inspired models of coordination emerged in the last decade as the most effective approaches to tackle the complexity of pervasive, intelligent, and self-* systems. In this talk we survey the most relevant nature-inspired coordination models, discuss the main open issues, and explore the potential for their future development.
[Invited Talk @ IDC 2012, Calabria, Italy, 26/9/2012]
Nature-inspired Coordination for Complex Multi-Agent SystemsAndrea Omicini
The document discusses nature-inspired coordination for complex multi-agent systems. It provides examples of early nature-inspired coordination models like stigmergy and chemical coordination. It also discusses modern field-based and (bio)chemical coordination models. It addresses basic issues with nature-inspired coordination regarding the environment and incorporating stochastic behavior. It discusses tuple-based coordination models and how Linda inspired many extensions, some of which are nature-inspired. It argues that fully capturing natural systems requires addressing time dependency and stochasticity in coordination models.
The TuCSoN Coordination Model & Technology. A GuideAndrea Omicini
The slides provide an introduction to the TuCSoN coordination model, and a guide to the TuCSoN coordination technology for distributed systems.
[Guide v. 1.3.1 for TuCSoN v. 1.12.0.0301]
Game Engines to Model MAS: A Research RoadmapAndrea Omicini
This document outlines a research roadmap for bridging the gap between multi-agent systems (MAS) and game engines (GE). It finds that while GEs excel at modeling environments, they provide weak support for agency and sociality as conceptualized in MAS. The roadmap proposes a "mirror worlds" approach where GEs represent environments and MAS handle autonomy and social interaction. A case study applies this by implementing the dining philosophers problem in Unity3D, with agents, chairs and a table modeled as game objects that coordinate through messaging. The roadmap concludes that pursuing GE-based MAS infrastructure could integrate the technologies' strengths while avoiding forcing either to handle responsibilities it is not meant for.
The document discusses using ontology to enhance management of operating system services. It proposes an ontology-based architecture to classify and organize various operating system services. The architecture divides services into the kernel and exokernel layers. Individual service ontologies are developed for search, backup, security, and networking and then merged into a universal ontology. This approach aims to improve service development, deployment and management by adding semantics and enabling knowledge sharing and reusability across operating systems.
The document summarizes and compares different methods for face recognition, including Eigenface, Line Edge Map (LEM), and other techniques. It provides descriptions of how each technique works, such as using eigenvectors to extract features for Eigenface. Experimental results show LEM achieves better accuracy than Eigenfaces for variations in lighting and size. While Eigenfaces struggles with size changes, LEM maintains high accuracy for different conditions. The document recommends future work combining techniques to maximize recognition accuracy.
Molecules of Knowledge: Self-Organisation in Knowledge-Intensive EnvironmentsAndrea Omicini
We propose a novel self-organising knowledge-oriented model based on biochemical tuple spaces, called Molecules of Knowledge (MoK). We introduce MoK basic entities, define its computational model, and discuss its mapping on the TuCSoN coordination model for its implementation.
[IDC 2012, Calabria, Italy, 24/9/2012
Autonomy, Interaction & Complexity in Computational Systems. Preliminary NotesAndrea Omicini
The issues of autonomy do not end with autonomous components. Autonomous systems are complex self-organising systems where interaction plays an essential role. Here we discuss some early thoughts on autonomy, interaction, and complexity in artificial systems, and suggest that nature-inspired coordination models should work as the main sources of technologies.
Nature-inspired Coordination for Complex Distributed SystemsAndrea Omicini
Originating from closed parallel systems, coordination models and technologies gained in expressive power so to deal with open distributed systems. In particular, nature-inspired models of coordination emerged in the last decade as the most effective approaches to tackle the complexity of pervasive, intelligent, and self-* systems. In this talk we survey the most relevant nature-inspired coordination models, discuss the main open issues, and explore the potential for their future development.
[Invited Talk @ IDC 2012, Calabria, Italy, 26/9/2012]
Nature-inspired Coordination for Complex Multi-Agent SystemsAndrea Omicini
The document discusses nature-inspired coordination for complex multi-agent systems. It provides examples of early nature-inspired coordination models like stigmergy and chemical coordination. It also discusses modern field-based and (bio)chemical coordination models. It addresses basic issues with nature-inspired coordination regarding the environment and incorporating stochastic behavior. It discusses tuple-based coordination models and how Linda inspired many extensions, some of which are nature-inspired. It argues that fully capturing natural systems requires addressing time dependency and stochasticity in coordination models.
The TuCSoN Coordination Model & Technology. A GuideAndrea Omicini
The slides provide an introduction to the TuCSoN coordination model, and a guide to the TuCSoN coordination technology for distributed systems.
[Guide v. 1.3.1 for TuCSoN v. 1.12.0.0301]
Game Engines to Model MAS: A Research RoadmapAndrea Omicini
This document outlines a research roadmap for bridging the gap between multi-agent systems (MAS) and game engines (GE). It finds that while GEs excel at modeling environments, they provide weak support for agency and sociality as conceptualized in MAS. The roadmap proposes a "mirror worlds" approach where GEs represent environments and MAS handle autonomy and social interaction. A case study applies this by implementing the dining philosophers problem in Unity3D, with agents, chairs and a table modeled as game objects that coordinate through messaging. The roadmap concludes that pursuing GE-based MAS infrastructure could integrate the technologies' strengths while avoiding forcing either to handle responsibilities it is not meant for.
This document discusses using ontologies as the basis for spatial data harmonization. It defines spatial data harmonization as the process of eliminating heterogeneities in spatial data properties to support interoperability. An ontology provides a formal representation of concepts and their relationships in a domain that can be used to describe data and map between different classifications and legends. The document proposes using an ontology to formalize the knowledge and rules for harmonizing land use/land cover data from different sources through an extraction, transformation, and loading tool.
Gradient-based Self-organisation Patterns of Anticipative AdaptationDanilo Pianini
The self-organisation Gradient pattern is known to be a key spatial data structure to make information local to its source become global knowledge, and to dynamically and adaptively steer agents to that source even in mobile and faulty environments – e.g. when obstacles unpredictably appear. In this paper we conceive new self-organisation mechanisms built upon this pattern to tackle anticipative adaptation. We ensure that the retrieval of a target of interest proactively reacts to locally-available information about future events, namely, the knowledge about future obstacles (e.g., expected jams or road interruption in a traffic control scenario) is used to emergently compute alternative and faster paths.
Introductory tutorial on the foundations of agents and multi-agent systems at the 18th European Agent Systems Summer School (EASSS 2016) – 25 July 2016, Catania, Italy
FInES, ENSEMBLE and A Scientific Perspective For Enterprise InteroperabilityFenareti Lampathaki
An overview of the interoperability history that led to the concept of the Enterprise Interoperability Science Base; the methodology followed; the main achievements so far and the key challenges towards the future.
From Coordination to Semantic Self-Organisation: A Perspective on the Enginee...Andrea Omicini
After briefly recapitulating the classical lines of the literature on coordination models, we discuss the new lines of research that aim at addressing the coordination of complex systems, then focus on mechanisms and patterns of coordination for self-organising systems. The notions of semantic coordination and self-organising coordination are defined and shortly discussed, then a vision of SOSC (self-organising semantic coordination) is presented, along with some insights over available technologies and possible scenarios for SOSC.
[Lecture at the PhD Mini-school, 11th National Workshop "From Objects to Agents" (WOA 2010) — 05/09/2010, Bologna, Italy.
Fundamentals of object orientation, objects, classes, classification and object models delivered to post-graduate students of Object Oriented Software Engineering.
Self-Organising News Management: The Molecules Of Knowledge ApproachStefano Mariani
This talk discusses the paper “Self-Organising News Management: The Molecules of Knowledge Approach”, presented at the 1st International Workshop on Adaptive Service Ecosystems: Natural and Socially Inspired Solutions (ASENSIS 2012).
Self-Organising News Management: The Molecules of Knowledge ApproachAndrea Omicini
Nowadays, news management systems present new critical challenges in the knowledge management process: the ever-increasing amount of information to handle, its hetero- geneity in structure, and the pace at which it is made available are just a few to mention. Features such as autonomy and self- organisation are apparently essential to face challenges of such a sort: we foresee systems where news are generated in shared spaces – compartments – as molecules of knowledge, which self- aggregate and autonomously move toward news prosumers— e.g., journalists. Along this line, we discuss the Molecules of Knowledge (MoK) model for self-organising news management, featuring biochemical tuple spaces for creation, aggregation, diffusion and consumption of news. We discuss the MoK general computational model and describe its main abstractions, then we focus on news management, showing how to integrate the state-of-art international standards for news representation and dissemination in MoK, thus leading to the MoK-News domain-specific model; finally we discuss our first experiments in self-organising knowledge-oriented coordination for news management.
[ASENSIS @ SASO 2012, Lyon, France — 10/09/2012]
This resume is for Seyedalireza Seyedi, who has a PhD in mathematics from Universita di Bologna. He has 7 years of experience in data science, machine learning, and optimization. His work includes large-scale data analysis projects at Universita di Bologna and algorithm development at Universiti Teknologi Malaysia. He is proficient in MATLAB, R, and other programming languages.
2D FEATURES-BASED DETECTOR AND DESCRIPTOR SELECTION SYSTEM FOR HIERARCHICAL R...gerogepatton
Detection and description of keypoints from an image is a well-studied problem in Computer Vision. Some methods like SIFT, SURF or ORB are computationally really efficient. This paper proposes a solution for a particular case study on object recognition of industrial parts based on hierarchical classification. Reducing the number of instances leads to better performance, indeed, that is what the use of the hierarchical classification is looking for. We demonstrate that this method performs better than using just one method
like ORB, SIFT or FREAK, despite being fairly slower.
2D FEATURES-BASED DETECTOR AND DESCRIPTOR SELECTION SYSTEM FOR HIERARCHICAL R...ijaia
This document proposes a hierarchical recognition system for industrial parts that selects the optimal 2D feature-based detection and description method based on the part characteristics. The system first classifies new parts into groups that behave similarly under different recognition pipelines. It then uses the pipeline that on average performs best for that group, improving overall accuracy compared to using a single pipeline. The proposed approach is evaluated using leave-one-out cross-validation to test different pipelines on labeled image data of the industrial parts.
2D Features-based Detector and Descriptor Selection System for Hierarchical R...gerogepatton
Detection and description of keypoints from an image is a well-studied problem in Computer Vision. Some
methods like SIFT, SURF or ORB are computationally really efficient. This paper proposes a solution for
a particular case study on object recognition of industrial parts based on hierarchical classification. Reducing the number of instances leads to better performance, indeed, that is what the use of the hierarchical
classification is looking for. We demonstrate that this method performs better than using just one method
like ORB, SIFT or FREAK, despite being fairly slower
Molecules Of Knowledge: Self-Organisation In Knowledge-Intensive EnvironmentsStefano Mariani
This talk discusses the paper “Molecules of Knowledge: Self-Organisation in Knowledge-Intensive Environments”, presented at the 6th International Symposium on Intelligent Distributed Computing (IDC 2012).
Open source software ecosystem modelling has emerged asan important research area in software engineering. Several models havebeen proposed to identify and analyse the complex relationships in OSS-ecosystems. However, there is a lack of formal models, methodologies,tool support, and standard notations for OSS-ecosystems. In this paperwe propose a general framework for support the OSS-ecosystems mod-elling process. This framework will allow the representation, synthesis,analysis, evaluation, and evolution of OSS-ecosystems. Design sciencemethodology is proposed to create several artefacts and investigatingthe suitability of these artefacts in the OSS-ecosystem context.
This document discusses an integrated approach to ontology development methodology and provides a case study using a shopping mall domain. It begins by reviewing existing ontology development methodologies and identifying their pitfalls. An integrated methodology is then proposed which aims to reduce these pitfalls. The key steps in the proposed methodology are: 1) capturing motivating user scenarios or keywords, 2) generating formal/informal questions and answers from the scenarios, 3) extracting terms and constraints, and 4) building the ontology using a top-down approach. The methodology is applied to developing an ontology for a shopping mall domain to provide multilingual information to visitors.
From Engineer to Alchemist, There and Back Again: An Alchemist TaleDanilo Pianini
The document discusses simulation and the Alchemist simulation tool. It begins with definitions of simulation and discusses how models and simulations are used. It then describes the computational model and engine architecture of Alchemist, which extends existing stochastic chemical simulation approaches. Several case studies using Alchemist are presented, including crowd evacuation and morphogenesis. Performance tests show Alchemist outperforms Repast for larger simulations. The development process of Alchemist is also distributed across multiple contributors.
Advanced Coordination Techniques: Experiments with TuCSoN and ReSpecTStefano Mariani
Distributed systems are more about interaction than computation: thus, they need coordination abstractions and techniques for managing the associated space of interaction.
Inspired to the archetypal Linda model, TuCSoN is a coordination model and technology providing tuple-based coordination services to (mobile) software agents through the notion of tuple centre—that is, a programmable tuple space. ReSpecT is a coordination language enabling run-time programmability of TuCSoN tuple centres—that is, of the coordination primitives and laws enabling and constraining interaction.
In this seminar TuCSoN and ReSpecT are used as the reference models and technologies for experimenting advanced coordination techniques for distributed and mobile programming of intelligent and pervasive multi-agent systems.
Feature location in a collection of software product variants using formal co...Ra'Fat Al-Msie'deen
This document presents an approach for locating features in a collection of software product variants using formal concept analysis (FCA). The approach abstracts each variant as a set of object-oriented building elements (OBEs) like packages, classes, methods. FCA is used to identify common features implemented across all variants as well as optional features implemented in some variants. The common features are identified as the top concept in the FCA concept lattice, while optional features are identified in other concepts. The approach was evaluated on variants of the ArgoUML modeling tool and correctly located all features.
This document contains definitions and explanations of object-oriented analysis and design concepts. It begins with definitions of polymorphism, the micro development process, object-oriented analysis and design, and the Unified Modeling Language (UML). It then discusses pattern templates, the need for a unified modeling approach, reasons for object orientation, the power of prototypes, multiple inheritance, and the goal of object-oriented design. The remaining sections provide more details on specific concepts like relationships, associations, frameworks, methodologies like Booch and design patterns.
Measuring Trustworthiness in Neuro-Symbolic IntegrationAndrea Omicini
This document discusses measuring trustworthiness in neuro-symbolic integration systems. It begins by providing background on neuro-symbolic integration systems and how they integrate neural and symbolic AI approaches. It then discusses the public and regulatory focus on ensuring the trustworthiness of AI systems. The document motivates the need to define metrics to measure how well neuro-symbolic systems meet requirements for trustworthy AI, as requirements alone may not be enough without metrics. It proposes exploring how to translate and measure compliance with requirements for areas like human oversight, robustness, and privacy in the context of neuro-symbolic systems.
Explainable Pervasive Intelligence with Self-explaining AgentsAndrea Omicini
Pervasiveness of ICT resources along with the promise of ubiquitous intelligence is pushing hard both our demand and our fears of AI: demand mandates for the ability to inject intelligence ubiquitously; fears compel the behaviour of intelligent systems to be observable, explainable, and accountable. Whereas the first wave of the new "AI Era" was mostly heralded by sub-symbolic approaches, features like explainability are better provided by symbolic techniques. In particular, the notion of explanation should be regarded as a core notion for intelligent systems, rather than just an add-on to make them understandable to humans. Based on symbolic AI techniques to match intuitive and rational cognition, explanation should then be regarded as a fundamental tool for inter-agent communication among heterogeneous intelligent agents in open multi-agent systems. More generally, self-explaining agents should work as the basic components in the engineering of intelligent systems integrating both symbolic and sub-/non-symbolic AI techniques.
More Related Content
Similar to Towards the Analysis & Prediction of Complex System Behaviour in SAPERE
This document discusses using ontologies as the basis for spatial data harmonization. It defines spatial data harmonization as the process of eliminating heterogeneities in spatial data properties to support interoperability. An ontology provides a formal representation of concepts and their relationships in a domain that can be used to describe data and map between different classifications and legends. The document proposes using an ontology to formalize the knowledge and rules for harmonizing land use/land cover data from different sources through an extraction, transformation, and loading tool.
Gradient-based Self-organisation Patterns of Anticipative AdaptationDanilo Pianini
The self-organisation Gradient pattern is known to be a key spatial data structure to make information local to its source become global knowledge, and to dynamically and adaptively steer agents to that source even in mobile and faulty environments – e.g. when obstacles unpredictably appear. In this paper we conceive new self-organisation mechanisms built upon this pattern to tackle anticipative adaptation. We ensure that the retrieval of a target of interest proactively reacts to locally-available information about future events, namely, the knowledge about future obstacles (e.g., expected jams or road interruption in a traffic control scenario) is used to emergently compute alternative and faster paths.
Introductory tutorial on the foundations of agents and multi-agent systems at the 18th European Agent Systems Summer School (EASSS 2016) – 25 July 2016, Catania, Italy
FInES, ENSEMBLE and A Scientific Perspective For Enterprise InteroperabilityFenareti Lampathaki
An overview of the interoperability history that led to the concept of the Enterprise Interoperability Science Base; the methodology followed; the main achievements so far and the key challenges towards the future.
From Coordination to Semantic Self-Organisation: A Perspective on the Enginee...Andrea Omicini
After briefly recapitulating the classical lines of the literature on coordination models, we discuss the new lines of research that aim at addressing the coordination of complex systems, then focus on mechanisms and patterns of coordination for self-organising systems. The notions of semantic coordination and self-organising coordination are defined and shortly discussed, then a vision of SOSC (self-organising semantic coordination) is presented, along with some insights over available technologies and possible scenarios for SOSC.
[Lecture at the PhD Mini-school, 11th National Workshop "From Objects to Agents" (WOA 2010) — 05/09/2010, Bologna, Italy.
Fundamentals of object orientation, objects, classes, classification and object models delivered to post-graduate students of Object Oriented Software Engineering.
Self-Organising News Management: The Molecules Of Knowledge ApproachStefano Mariani
This talk discusses the paper “Self-Organising News Management: The Molecules of Knowledge Approach”, presented at the 1st International Workshop on Adaptive Service Ecosystems: Natural and Socially Inspired Solutions (ASENSIS 2012).
Self-Organising News Management: The Molecules of Knowledge ApproachAndrea Omicini
Nowadays, news management systems present new critical challenges in the knowledge management process: the ever-increasing amount of information to handle, its hetero- geneity in structure, and the pace at which it is made available are just a few to mention. Features such as autonomy and self- organisation are apparently essential to face challenges of such a sort: we foresee systems where news are generated in shared spaces – compartments – as molecules of knowledge, which self- aggregate and autonomously move toward news prosumers— e.g., journalists. Along this line, we discuss the Molecules of Knowledge (MoK) model for self-organising news management, featuring biochemical tuple spaces for creation, aggregation, diffusion and consumption of news. We discuss the MoK general computational model and describe its main abstractions, then we focus on news management, showing how to integrate the state-of-art international standards for news representation and dissemination in MoK, thus leading to the MoK-News domain-specific model; finally we discuss our first experiments in self-organising knowledge-oriented coordination for news management.
[ASENSIS @ SASO 2012, Lyon, France — 10/09/2012]
This resume is for Seyedalireza Seyedi, who has a PhD in mathematics from Universita di Bologna. He has 7 years of experience in data science, machine learning, and optimization. His work includes large-scale data analysis projects at Universita di Bologna and algorithm development at Universiti Teknologi Malaysia. He is proficient in MATLAB, R, and other programming languages.
2D FEATURES-BASED DETECTOR AND DESCRIPTOR SELECTION SYSTEM FOR HIERARCHICAL R...gerogepatton
Detection and description of keypoints from an image is a well-studied problem in Computer Vision. Some methods like SIFT, SURF or ORB are computationally really efficient. This paper proposes a solution for a particular case study on object recognition of industrial parts based on hierarchical classification. Reducing the number of instances leads to better performance, indeed, that is what the use of the hierarchical classification is looking for. We demonstrate that this method performs better than using just one method
like ORB, SIFT or FREAK, despite being fairly slower.
2D FEATURES-BASED DETECTOR AND DESCRIPTOR SELECTION SYSTEM FOR HIERARCHICAL R...ijaia
This document proposes a hierarchical recognition system for industrial parts that selects the optimal 2D feature-based detection and description method based on the part characteristics. The system first classifies new parts into groups that behave similarly under different recognition pipelines. It then uses the pipeline that on average performs best for that group, improving overall accuracy compared to using a single pipeline. The proposed approach is evaluated using leave-one-out cross-validation to test different pipelines on labeled image data of the industrial parts.
2D Features-based Detector and Descriptor Selection System for Hierarchical R...gerogepatton
Detection and description of keypoints from an image is a well-studied problem in Computer Vision. Some
methods like SIFT, SURF or ORB are computationally really efficient. This paper proposes a solution for
a particular case study on object recognition of industrial parts based on hierarchical classification. Reducing the number of instances leads to better performance, indeed, that is what the use of the hierarchical
classification is looking for. We demonstrate that this method performs better than using just one method
like ORB, SIFT or FREAK, despite being fairly slower
Molecules Of Knowledge: Self-Organisation In Knowledge-Intensive EnvironmentsStefano Mariani
This talk discusses the paper “Molecules of Knowledge: Self-Organisation in Knowledge-Intensive Environments”, presented at the 6th International Symposium on Intelligent Distributed Computing (IDC 2012).
Open source software ecosystem modelling has emerged asan important research area in software engineering. Several models havebeen proposed to identify and analyse the complex relationships in OSS-ecosystems. However, there is a lack of formal models, methodologies,tool support, and standard notations for OSS-ecosystems. In this paperwe propose a general framework for support the OSS-ecosystems mod-elling process. This framework will allow the representation, synthesis,analysis, evaluation, and evolution of OSS-ecosystems. Design sciencemethodology is proposed to create several artefacts and investigatingthe suitability of these artefacts in the OSS-ecosystem context.
This document discusses an integrated approach to ontology development methodology and provides a case study using a shopping mall domain. It begins by reviewing existing ontology development methodologies and identifying their pitfalls. An integrated methodology is then proposed which aims to reduce these pitfalls. The key steps in the proposed methodology are: 1) capturing motivating user scenarios or keywords, 2) generating formal/informal questions and answers from the scenarios, 3) extracting terms and constraints, and 4) building the ontology using a top-down approach. The methodology is applied to developing an ontology for a shopping mall domain to provide multilingual information to visitors.
From Engineer to Alchemist, There and Back Again: An Alchemist TaleDanilo Pianini
The document discusses simulation and the Alchemist simulation tool. It begins with definitions of simulation and discusses how models and simulations are used. It then describes the computational model and engine architecture of Alchemist, which extends existing stochastic chemical simulation approaches. Several case studies using Alchemist are presented, including crowd evacuation and morphogenesis. Performance tests show Alchemist outperforms Repast for larger simulations. The development process of Alchemist is also distributed across multiple contributors.
Advanced Coordination Techniques: Experiments with TuCSoN and ReSpecTStefano Mariani
Distributed systems are more about interaction than computation: thus, they need coordination abstractions and techniques for managing the associated space of interaction.
Inspired to the archetypal Linda model, TuCSoN is a coordination model and technology providing tuple-based coordination services to (mobile) software agents through the notion of tuple centre—that is, a programmable tuple space. ReSpecT is a coordination language enabling run-time programmability of TuCSoN tuple centres—that is, of the coordination primitives and laws enabling and constraining interaction.
In this seminar TuCSoN and ReSpecT are used as the reference models and technologies for experimenting advanced coordination techniques for distributed and mobile programming of intelligent and pervasive multi-agent systems.
Feature location in a collection of software product variants using formal co...Ra'Fat Al-Msie'deen
This document presents an approach for locating features in a collection of software product variants using formal concept analysis (FCA). The approach abstracts each variant as a set of object-oriented building elements (OBEs) like packages, classes, methods. FCA is used to identify common features implemented across all variants as well as optional features implemented in some variants. The common features are identified as the top concept in the FCA concept lattice, while optional features are identified in other concepts. The approach was evaluated on variants of the ArgoUML modeling tool and correctly located all features.
This document contains definitions and explanations of object-oriented analysis and design concepts. It begins with definitions of polymorphism, the micro development process, object-oriented analysis and design, and the Unified Modeling Language (UML). It then discusses pattern templates, the need for a unified modeling approach, reasons for object orientation, the power of prototypes, multiple inheritance, and the goal of object-oriented design. The remaining sections provide more details on specific concepts like relationships, associations, frameworks, methodologies like Booch and design patterns.
Similar to Towards the Analysis & Prediction of Complex System Behaviour in SAPERE (20)
Measuring Trustworthiness in Neuro-Symbolic IntegrationAndrea Omicini
This document discusses measuring trustworthiness in neuro-symbolic integration systems. It begins by providing background on neuro-symbolic integration systems and how they integrate neural and symbolic AI approaches. It then discusses the public and regulatory focus on ensuring the trustworthiness of AI systems. The document motivates the need to define metrics to measure how well neuro-symbolic systems meet requirements for trustworthy AI, as requirements alone may not be enough without metrics. It proposes exploring how to translate and measure compliance with requirements for areas like human oversight, robustness, and privacy in the context of neuro-symbolic systems.
Explainable Pervasive Intelligence with Self-explaining AgentsAndrea Omicini
Pervasiveness of ICT resources along with the promise of ubiquitous intelligence is pushing hard both our demand and our fears of AI: demand mandates for the ability to inject intelligence ubiquitously; fears compel the behaviour of intelligent systems to be observable, explainable, and accountable. Whereas the first wave of the new "AI Era" was mostly heralded by sub-symbolic approaches, features like explainability are better provided by symbolic techniques. In particular, the notion of explanation should be regarded as a core notion for intelligent systems, rather than just an add-on to make them understandable to humans. Based on symbolic AI techniques to match intuitive and rational cognition, explanation should then be regarded as a fundamental tool for inter-agent communication among heterogeneous intelligent agents in open multi-agent systems. More generally, self-explaining agents should work as the basic components in the engineering of intelligent systems integrating both symbolic and sub-/non-symbolic AI techniques.
On the Integration of Symbolic and Sub-symbolic – Explaining by DesignAndrea Omicini
The more intelligent systems based on sub-symbolic techniques pervade our everyday lives, the less human can understand them. This is why symbolic approaches are getting more and more attention in the general effort to make AI interpretable, explainable, and trustable. Understanding the current state of the art of AI techniques integrating symbolic and sub-symbolic approaches is then of paramount importance, nowadays—in particular in the XAI perspective. In this talk we first provides an overview of the main symbolic/sub-symbolic integration techniques, focussing in particular on those targeting explainable AI systems. Then we expand the notion of “explainability by design” to the realm of multi-agent systems, where XAI techniques can play a key role in the engineering of intelligent systems.
Not just for humans: Explanation for agent-to-agent communicationAndrea Omicini
Once precisely defined so as to include just the explanation’s act, the notion of explanation should be regarded as a central notion in the engineering of intelligent system—not just as an add-on to make them understandable to humans. Based on symbolic AI techniques to match intuitive and rational cognition, explanation should be exploited as a fundamental tool for inter-agent communication among heterogeneous agents in open multi-agent systems. More generally, explanation-ready agents should work as the basic components in the engineering of intelligent systems integrating both symbolic and sub-/non-symbolic AI techniques.
Presented by Andrea Omicini @ AIxIA 2020 Discussion Paper Workshop
Blockchain for Intelligent Systems: Research PerspectivesAndrea Omicini
We summarise and compare features of MAS and BCT, and discuss how they could be fruitfully integrated in the engineering of intelligent systems by adopting a long-term research perspective.
Injecting (Micro)Intelligence in the IoT: Logic-based Approaches for (M)MASAndrea Omicini
Pervasiveness of ICT resources along with the promise of ubiquitous intelligence is pushing hard both our demand and our fears of AI: demand mandates for the ability to inject (micro) intelligence ubiquitously, fears compel the behaviour of intelligent systems to be observable, explainable, and accountable.
Whereas the first wave of the new "AI Era" was mostly heralded by non-symbolic approaches, features like explainability are better provided by symbolic techniques.
In this talk we focus on logic-based approaches, and discuss their potential in pervasive scenarios like the IoT and open (M)MAS along with our latest results in the field.
Andrea Omicini, Roberta Calegari
Invited Talk
MMAS 2018, Stockholm, Sweden, 14 July 2018
Conversational Informatics: From Conversational Systems to Communication Inte...Andrea Omicini
The document discusses the history and development of conversational systems. It outlines some of the early work in natural language dialogue systems from the 1960s and focuses on three main areas of research: understanding and generating stories, cognitive systems, and integration of these approaches. Current conversational systems aim to more fully integrate these areas to provide more human-like conversational abilities.
Complexity in computational systems: the coordination perspectiveAndrea Omicini
In this talk we discuss the role of coordination models and technologies in the engineering of complex computational systems.
Complex Systems Physics Meeting IMT-UNIBO
Dipartimento di Fisica e Astronomia, Università di Bologna
Bologna, Italy, 15/02/2018
Nature-inspired Coordination: Current Status and Research TrendsAndrea Omicini
Tutorial @WI 2017, Leipzig, 23 August 2017
Andrea Omicini & Stefano Mariani, Lecturers
Originating from closed parallel systems, coordination models and technologies gained in expressive power so as to deal with complex distributed systems. In particular, nature-inspired models of coordination emerged in the last decade as the most effective approaches to tackle the complexity of pervasive, intelligent, and self-* systems.
In the first part of the tutorial we introduce the basic notions of coordination and coordination model, and relate them to the notions of interaction and complexity. Then, the most relevant nature-inspired coordination (NIC) models are discussed, along with their relationship with the many facets of tuple-based models. In the third part we discuss the main open issues and explore the trends for future development of NIC. Finally, as a case study, we focus on MoK (Molecules of Knowledge), a NIC model for knowledge self-organisation, where data and information autonomously aggregate and spread toward knowledge prosumers.
Novel Opportunities for Tuple-based Coordination: XPath, the Blockchain, and ...Andrea Omicini
The increasing maturity of some well-established technologies – such as XPath – along with the sharp rise of brand-new ones – i.e. the blockchain – presents new opportunities to researchers in the field of multi-agent coordination. In this talk we briefly discuss a few technologies which, once suitably interpreted and integrated, have the potential to impact the very roots of tuple-based coordination as it stems from the archetypal LINDA model.
Micro-intelligence for the IoT: Teaching the Old Logic Dog New Programming Tr...Andrea Omicini
New application scenarios for pervasive intelligent systems open novel perspectives for logic-based approaches, in particular when coupled with agent-based technologies and methods. In this explorative talk we provide some examples of how logic programming and its extensions can work as sources of micro-intelligence for the IoT, at both the individual and the collective level, along with an overall architectural view of IoT systems exploiting logic-based technologies.
Logic Programming as a Service (LPaaS): Intelligence for the IoTAndrea Omicini
Talk @ ICNSC 2017, Calabria, Italy, 16 May 2017
Abstract: The widespread diffusion of low-cost computing devices, such as Arduino boards and Raspberry Pi, along with improvements of Cloud computing platforms, are paving the way towards a whole new set of opportunities for Internet of Things (IoT) applications and services. Varying degrees of intelligence are often required for supporting adaptation and self-management—yet, they should be provided in a light-weight, easy to use and customise, highly-interoperable way. Accordingly, in this paper we explore the idea of Logic Programming as a Service (LPaaS) as a novel and promising re-interpretation of distributed logic programming in the IoT era. After introducing the reference context and motivating scenarios of LPaaS as a key enabling technology for intelligent IoT, we define the LPaaS general system architecture. Then, we present a prototype implementation built on top of the tuProlog system, which provides the required interoperability and customisation. We showcase the LPaaS potential through a case study designed as a simplification of the motivating scenarios.
Privacy through Anonymisation in Large-scale Socio-technical Systems: The BIS...Andrea Omicini
Large-scale socio-technical systems (STS) inextricably inter-connect individual – e.g., the right to privacy –, social – e.g., the effectiveness of organisational processes –, and technology issues —e.g., the software engineering process. As a result, the design of the complex software infrastructure involves also non-technological aspects such as the legal ones—so that, e.g., law-abidingness can be ensured since the early stages of the software engineering process. By focussing on contact centres (CC) as relevant examples of knowledge-intensive STS, we elaborate on the articulate aspects of anonymisation: there, individual and organisational needs clash, so that only an accurate balancing between legal and technical aspects could possibly ensure the system efficiency while preserving the individual right to privacy. We discuss first the overall legal framework, then the general theme of anonymisation in CC. Finally we overview the technical process developed in the context of the BISON project.
Project presentation @ DMI, Università di Catania, Italy, 25 July 2016
The impact of mobile technologies on healthcare is particularly evident in the case of self-management of chronic diseases, where they can decrease spending and improve the patient quality of life. In this talk we propose the adoption of agent-based modelling and simulation techniques as built-in tools to dynamically monitor patient health state and provide feedbacks for self-management. To demonstrate the feasibility of our proposal we focus on Type 1 Diabetes Mellitus as our case study, and provide some preliminary simulation results.
Open distributed multi-agent systems featuring autonomous components demand coordination mechanisms for both functional and non-functional properties. Heterogeneity of requirements regarding interaction means and paradigms, stemming from the diverse nature of components, should not affect the effectiveness of coordination. Along this line, in this paper we share our pragmatical experience in the integration of objective and subjective, synchronous and asynchronous, reactive and pro-active coordination approaches within two widely-adopted agent-oriented technologies (JADE and Jason), enabling coordinating components to dynamically adapt their interaction means based on static preference or run-time contingencies.
Towards Logic Programming as a Service: Experiments in tuPrologAndrea Omicini
The document discusses experiments with logic programming as a service (LPaaS) using tuProlog. It proposes an LPaaS architecture with local APIs for creating engines, loading theories, and querying. It also describes implementing tuProlog as a service (2PaaS) on iOS, where tuProlog is embedded in an app and provides an API via URL schemes for other apps to interface with it. Example application scenarios are presented, including using tuProlog for symbolic derivatives and multi-language programming. Further work is outlined to better define and extend the LPaaS paradigm.
The document discusses the concept of space in computer science and computational systems. It describes how physical distribution of computational units and communication channels creates virtual spaces, such as distributed systems and middleware environments. It emphasizes that situated distributed systems and pervasive computing systems are needed to handle computation in physical environments and take into account the spatial aspects of those environments. Situatedness and context-awareness are important properties for computational systems to exhibit as they become more immersed in physical space.
Academic Publishing in the Digital Era: A Couple of Issues (Open Access—Well,...Andrea Omicini
The document discusses issues with the current academic publishing system. It notes that while technology has made production, diffusion, and access to scholarly works negligible in cost, access to most literature remains expensive. It also distinguishes between the validation of research, which ensures quality, and evaluation, which occurs after publication. While new publishers and self-archiving aim to address costs, true validation requires peer review before rather than after publication to maintain standards and scale effectively. Overall reforms are needed to make the system more economically sustainable while maintaining scientific rigor.
Self-organisation of Knowledge in Socio-technical Systems: A Coordination Per...Andrea Omicini
Some of the most peculiar traits of socio-technical systems (STS) in knowledge-intensive environments (KIE) – such as unpredictability of agents’ behaviour, ever-growing amount of information to manage, fast-paced production/consumption – tangle coordination of agents as well as coordination of information, by affecting, e.g., reachability by knowledge prosumers and manageability by the IT infrastructure. In this seminar we describe a novel approach to coordination of STS in KIE, grounded on the MoK (Molecules of Knowledge) model for knowledge self-organisation, and inspired to key concepts from the cognitive theory of BIC (behavioural implicit communication).
Anticipatory Coordination in Socio-technical Knowledge-intensive Environments...Andrea Omicini
This document discusses using behavioral implicit communication (BIC) to enable anticipatory coordination in socio-technical knowledge-intensive environments. It proposes extending the Molecules of Knowledge (MoK) coordination model to treat computational environments as shared environments where agents' actions and traces are observable. Enzymes and traces in MoK are redefined to represent epistemic actions and their side effects, enabling the environment to implicitly understand intentions and goals. This lays the groundwork for collective intelligence and anticipatory coordination through the tacit messages conveyed by agents' observable behaviors. An experiment simulates a citizen journalism scenario where search actions in a MoK-based system spread information in a self-organizing, adaptive manner.
Your One-Stop Shop for Python Success: Top 10 US Python Development Providersakankshawande
Simplify your search for a reliable Python development partner! This list presents the top 10 trusted US providers offering comprehensive Python development services, ensuring your project's success from conception to completion.
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdfMalak Abu Hammad
Discover how MongoDB Atlas and vector search technology can revolutionize your application's search capabilities. This comprehensive presentation covers:
* What is Vector Search?
* Importance and benefits of vector search
* Practical use cases across various industries
* Step-by-step implementation guide
* Live demos with code snippets
* Enhancing LLM capabilities with vector search
* Best practices and optimization strategies
Perfect for developers, AI enthusiasts, and tech leaders. Learn how to leverage MongoDB Atlas to deliver highly relevant, context-aware search results, transforming your data retrieval process. Stay ahead in tech innovation and maximize the potential of your applications.
#MongoDB #VectorSearch #AI #SemanticSearch #TechInnovation #DataScience #LLM #MachineLearning #SearchTechnology
Ocean lotus Threat actors project by John Sitima 2024 (1).pptxSitimaJohn
Ocean Lotus cyber threat actors represent a sophisticated, persistent, and politically motivated group that poses a significant risk to organizations and individuals in the Southeast Asian region. Their continuous evolution and adaptability underscore the need for robust cybersecurity measures and international cooperation to identify and mitigate the threats posed by such advanced persistent threat groups.
UiPath Test Automation using UiPath Test Suite series, part 6DianaGray10
Welcome to UiPath Test Automation using UiPath Test Suite series part 6. In this session, we will cover Test Automation with generative AI and Open AI.
UiPath Test Automation with generative AI and Open AI webinar offers an in-depth exploration of leveraging cutting-edge technologies for test automation within the UiPath platform. Attendees will delve into the integration of generative AI, a test automation solution, with Open AI advanced natural language processing capabilities.
Throughout the session, participants will discover how this synergy empowers testers to automate repetitive tasks, enhance testing accuracy, and expedite the software testing life cycle. Topics covered include the seamless integration process, practical use cases, and the benefits of harnessing AI-driven automation for UiPath testing initiatives. By attending this webinar, testers, and automation professionals can gain valuable insights into harnessing the power of AI to optimize their test automation workflows within the UiPath ecosystem, ultimately driving efficiency and quality in software development processes.
What will you get from this session?
1. Insights into integrating generative AI.
2. Understanding how this integration enhances test automation within the UiPath platform
3. Practical demonstrations
4. Exploration of real-world use cases illustrating the benefits of AI-driven test automation for UiPath
Topics covered:
What is generative AI
Test Automation with generative AI and Open AI.
UiPath integration with generative AI
Speaker:
Deepak Rai, Automation Practice Lead, Boundaryless Group and UiPath MVP
Infrastructure Challenges in Scaling RAG with Custom AI modelsZilliz
Building Retrieval-Augmented Generation (RAG) systems with open-source and custom AI models is a complex task. This talk explores the challenges in productionizing RAG systems, including retrieval performance, response synthesis, and evaluation. We’ll discuss how to leverage open-source models like text embeddings, language models, and custom fine-tuned models to enhance RAG performance. Additionally, we’ll cover how BentoML can help orchestrate and scale these AI components efficiently, ensuring seamless deployment and management of RAG systems in the cloud.
CAKE: Sharing Slices of Confidential Data on BlockchainClaudio Di Ciccio
Presented at the CAiSE 2024 Forum, Intelligent Information Systems, June 6th, Limassol, Cyprus.
Synopsis: Cooperative information systems typically involve various entities in a collaborative process within a distributed environment. Blockchain technology offers a mechanism for automating such processes, even when only partial trust exists among participants. The data stored on the blockchain is replicated across all nodes in the network, ensuring accessibility to all participants. While this aspect facilitates traceability, integrity, and persistence, it poses challenges for adopting public blockchains in enterprise settings due to confidentiality issues. In this paper, we present a software tool named Control Access via Key Encryption (CAKE), designed to ensure data confidentiality in scenarios involving public blockchains. After outlining its core components and functionalities, we showcase the application of CAKE in the context of a real-world cyber-security project within the logistics domain.
Paper: https://doi.org/10.1007/978-3-031-61000-4_16
Monitoring and Managing Anomaly Detection on OpenShift.pdfTosin Akinosho
Monitoring and Managing Anomaly Detection on OpenShift
Overview
Dive into the world of anomaly detection on edge devices with our comprehensive hands-on tutorial. This SlideShare presentation will guide you through the entire process, from data collection and model training to edge deployment and real-time monitoring. Perfect for those looking to implement robust anomaly detection systems on resource-constrained IoT/edge devices.
Key Topics Covered
1. Introduction to Anomaly Detection
- Understand the fundamentals of anomaly detection and its importance in identifying unusual behavior or failures in systems.
2. Understanding Edge (IoT)
- Learn about edge computing and IoT, and how they enable real-time data processing and decision-making at the source.
3. What is ArgoCD?
- Discover ArgoCD, a declarative, GitOps continuous delivery tool for Kubernetes, and its role in deploying applications on edge devices.
4. Deployment Using ArgoCD for Edge Devices
- Step-by-step guide on deploying anomaly detection models on edge devices using ArgoCD.
5. Introduction to Apache Kafka and S3
- Explore Apache Kafka for real-time data streaming and Amazon S3 for scalable storage solutions.
6. Viewing Kafka Messages in the Data Lake
- Learn how to view and analyze Kafka messages stored in a data lake for better insights.
7. What is Prometheus?
- Get to know Prometheus, an open-source monitoring and alerting toolkit, and its application in monitoring edge devices.
8. Monitoring Application Metrics with Prometheus
- Detailed instructions on setting up Prometheus to monitor the performance and health of your anomaly detection system.
9. What is Camel K?
- Introduction to Camel K, a lightweight integration framework built on Apache Camel, designed for Kubernetes.
10. Configuring Camel K Integrations for Data Pipelines
- Learn how to configure Camel K for seamless data pipeline integrations in your anomaly detection workflow.
11. What is a Jupyter Notebook?
- Overview of Jupyter Notebooks, an open-source web application for creating and sharing documents with live code, equations, visualizations, and narrative text.
12. Jupyter Notebooks with Code Examples
- Hands-on examples and code snippets in Jupyter Notebooks to help you implement and test anomaly detection models.
GraphRAG for Life Science to increase LLM accuracyTomaz Bratanic
GraphRAG for life science domain, where you retriever information from biomedical knowledge graphs using LLMs to increase the accuracy and performance of generated answers
For the full video of this presentation, please visit: https://www.edge-ai-vision.com/2024/06/building-and-scaling-ai-applications-with-the-nx-ai-manager-a-presentation-from-network-optix/
Robin van Emden, Senior Director of Data Science at Network Optix, presents the “Building and Scaling AI Applications with the Nx AI Manager,” tutorial at the May 2024 Embedded Vision Summit.
In this presentation, van Emden covers the basics of scaling edge AI solutions using the Nx tool kit. He emphasizes the process of developing AI models and deploying them globally. He also showcases the conversion of AI models and the creation of effective edge AI pipelines, with a focus on pre-processing, model conversion, selecting the appropriate inference engine for the target hardware and post-processing.
van Emden shows how Nx can simplify the developer’s life and facilitate a rapid transition from concept to production-ready applications.He provides valuable insights into developing scalable and efficient edge AI solutions, with a strong focus on practical implementation.
Removing Uninteresting Bytes in Software FuzzingAftab Hussain
Imagine a world where software fuzzing, the process of mutating bytes in test seeds to uncover hidden and erroneous program behaviors, becomes faster and more effective. A lot depends on the initial seeds, which can significantly dictate the trajectory of a fuzzing campaign, particularly in terms of how long it takes to uncover interesting behaviour in your code. We introduce DIAR, a technique designed to speedup fuzzing campaigns by pinpointing and eliminating those uninteresting bytes in the seeds. Picture this: instead of wasting valuable resources on meaningless mutations in large, bloated seeds, DIAR removes the unnecessary bytes, streamlining the entire process.
In this work, we equipped AFL, a popular fuzzer, with DIAR and examined two critical Linux libraries -- Libxml's xmllint, a tool for parsing xml documents, and Binutil's readelf, an essential debugging and security analysis command-line tool used to display detailed information about ELF (Executable and Linkable Format). Our preliminary results show that AFL+DIAR does not only discover new paths more quickly but also achieves higher coverage overall. This work thus showcases how starting with lean and optimized seeds can lead to faster, more comprehensive fuzzing campaigns -- and DIAR helps you find such seeds.
- These are slides of the talk given at IEEE International Conference on Software Testing Verification and Validation Workshop, ICSTW 2022.
Best 20 SEO Techniques To Improve Website Visibility In SERPPixlogix Infotech
Boost your website's visibility with proven SEO techniques! Our latest blog dives into essential strategies to enhance your online presence, increase traffic, and rank higher on search engines. From keyword optimization to quality content creation, learn how to make your site stand out in the crowded digital landscape. Discover actionable tips and expert insights to elevate your SEO game.
In the rapidly evolving landscape of technologies, XML continues to play a vital role in structuring, storing, and transporting data across diverse systems. The recent advancements in artificial intelligence (AI) present new methodologies for enhancing XML development workflows, introducing efficiency, automation, and intelligent capabilities. This presentation will outline the scope and perspective of utilizing AI in XML development. The potential benefits and the possible pitfalls will be highlighted, providing a balanced view of the subject.
We will explore the capabilities of AI in understanding XML markup languages and autonomously creating structured XML content. Additionally, we will examine the capacity of AI to enrich plain text with appropriate XML markup. Practical examples and methodological guidelines will be provided to elucidate how AI can be effectively prompted to interpret and generate accurate XML markup.
Further emphasis will be placed on the role of AI in developing XSLT, or schemas such as XSD and Schematron. We will address the techniques and strategies adopted to create prompts for generating code, explaining code, or refactoring the code, and the results achieved.
The discussion will extend to how AI can be used to transform XML content. In particular, the focus will be on the use of AI XPath extension functions in XSLT, Schematron, Schematron Quick Fixes, or for XML content refactoring.
The presentation aims to deliver a comprehensive overview of AI usage in XML development, providing attendees with the necessary knowledge to make informed decisions. Whether you’re at the early stages of adopting AI or considering integrating it in advanced XML development, this presentation will cover all levels of expertise.
By highlighting the potential advantages and challenges of integrating AI with XML development tools and languages, the presentation seeks to inspire thoughtful conversation around the future of XML development. We’ll not only delve into the technical aspects of AI-powered XML development but also discuss practical implications and possible future directions.
Have you ever been confused by the myriad of choices offered by AWS for hosting a website or an API?
Lambda, Elastic Beanstalk, Lightsail, Amplify, S3 (and more!) can each host websites + APIs. But which one should we choose?
Which one is cheapest? Which one is fastest? Which one will scale to meet our needs?
Join me in this session as we dive into each AWS hosting service to determine which one is best for your scenario and explain why!
Choosing The Best AWS Service For Your Website + API.pptx
Towards the Analysis & Prediction of Complex System Behaviour in SAPERE
1. Towards the Analysis & Prediction of Complex System
Behaviour in SAPERE
Marco Alberti & Andrea Omicini
marco.alberti3@studio.unibo.it, andrea.omicini@unibo.it
Alma Mater Studiorum—Universit` di Bologna
a
ASENSIS 2012
1st International Workshop @ SASO 2012,
Lyon, France, 10th of September 2012
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 1 / 36
2. Outline
1 Motivations
2 Steps
3 Background
4 SAPERE
5 Model-checking: A Possible Approach
6 Data Mining: A Possible Approach
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 2 / 36
3. Motivations
Outline
1 Motivations
2 Steps
3 Background
4 SAPERE
5 Model-checking: A Possible Approach
6 Data Mining: A Possible Approach
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 3 / 36
4. Motivations
Complexity & Unpredictability
Complex computational systems are typically unpredictable
for both practical & theoretical reasons
such as their situatedness within unpredictable environment, or the
non-compositionality of their inner components
Predictability is anyway an issue
again, for both theoretical & practical reasons
our ability to model complexity defines more or less the level of our
scientific achievements
our ability to use models to predict behaviours is an essential premise
to the engineering of artificial systems
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 4 / 36
5. Motivations
SAPERE Complex Systems
SAPERE systems are complex
nature-inspired, adaptive, pervasive, self-organising, multi-agent
systems
SAPERE models are articulated
SAPERE methodology, too
analysis & prediction tools are required, to be integrated in the
SAPERE SE process
SAPERE systems are coordinated systems
. . . so, we could have taken that route. . .
. . . instead, we choose to forget about it for a while – actually, it the
subject of another research stream – and to focus on the many
approaches available in the literature
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 5 / 36
6. Steps
Outline
1 Motivations
2 Steps
3 Background
4 SAPERE
5 Model-checking: A Possible Approach
6 Data Mining: A Possible Approach
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 6 / 36
7. Steps
Survey
Analysis and prediction of complex computational systems
we survey the most relevant approaches
we focus on those “reducible” to the SAPERE systems, in some ways
Analysis and prediction of SAPERE computational systems
we identify the two most promising techniques: model checking and
data mining
we derive two potentially-effective approaches to the analysis and
prediction of the behaviour of SAPERE pervasive ecosystems
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 7 / 36
8. Background
Outline
1 Motivations
2 Steps
3 Background
4 SAPERE
5 Model-checking: A Possible Approach
6 Data Mining: A Possible Approach
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 8 / 36
9. Background
MAS & Pervasive Systems
Many approaches, limited results
The most effective approaches deal with very narrow portions of the
system behaviour, perform quite limited sorts of analysis, and are
typically applicable to quite specific domains
A possible classification for MAS
a priori approaches dealing with compliance to the rules steering
system evolution
a posteriori approaches aimed at predicting system behaviour focusing on
specific system features
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 9 / 36
10. Background
Analysis by Model Checking I
Focus
(semi-)exhaustive analysis of the states the system will go through,
typically along with by a representation by a LTS (Labelled Transition
System)
verification of system meeting given properties, expressed though
various logics (LTL, CTL, DLTL, GLTL, etc.)
Uses
mostly used for the verification of safety or liveness properties
quite effective for “paper” problems, its application is however
complex for generic analysis, and often prohibitive from the
computational point of view
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 10 / 36
11. Background
Analysis by Model Checking II
Logics
LTL is usually preferred to CTL
From LTL numerous extensions have arisen, including
DLTL (Dynamic Linear Temporal Logic) focussed on enhancing the until
operator to better express pre- and post-conditions
[Giordano et al., 2004]
GLTL (General Linear Temporal Logic) allows the definition of LTL properties
for the whole system, as well as for the individual entities involved
[Pokorny and Ramakrishnan, 2006]
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 11 / 36
12. Background
Analysis by Model Checking III
BDI agents
BDI specifications can be translated into LTL specifications
AgentSpeak(L) dominates among BDI languages, and
AgentSpeak(F) is a subset of AgentSpeak(L), for finite state
systems only
[Bordini et al., 2003, Bordini et al., 2004, Bordini et al., 2006]
a system described by AgentSpeak(F) can be represented by a SLPN
(Simple Logic Petri Net), and model-checked [Behrens and Dix, 2007]
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 12 / 36
13. Background
Analysis of Swarm Systems
Homogeneous systems
a detailed model checking of individual entities is unthinkable because
of the explosion in the number of states
under the hypothesis of a system composed of homogeneous entities
[Konur et al., 2010, Dixon et al., 2011], it is possible to define a state
diagram common to all entities
a single LTS can be used to represent all the entities, by assigning a
number to each admissible state for the individual entities, which
represents at any instant of time the number of the system entities
currently in that particular state
each state is then associated to the number of entities that actually
are in that state at any given instant of time, using probabilistic
analysis
such a number is computed by using probabilistic analysis, so it
represents an estimation of the real number of entities that will be in
the state at runtime
however, noise tolerance of swarm systems ensures the reliability of
the analysis
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 13 / 36
14. Background
Analysis by Data Mining I
A posteriori
given MAS complexity – and opaqueness of agents – a posteriori
analysis of the behaviour is often attempted
the most satisfactory techniques exploit data mining algorithms to
analyse past behaviours of systems, to understand the rules actually
governing the system evolution
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 14 / 36
15. Background
Analysis by Data Mining II
Examples
a technique aimed at reducing the complexity of the behaviour of
RoboCup team down to atomic actions is presented in
[Kaminka et al., 2003]
good results based on the same sort are reported in
[Lattner et al., 2006] for strategy prediction in football teams of
robots
in [Mnif et al., 2007], a similar approach was used to identify the
different phases of the behaviour of a swarm system modelling the
interactions between chickens inside their cages
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 15 / 36
16. SAPERE
Outline
1 Motivations
2 Steps
3 Background
4 SAPERE
5 Model-checking: A Possible Approach
6 Data Mining: A Possible Approach
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 16 / 36
17. SAPERE
Self-aware Pervasive Service Ecosystems I
SAPERE project
European Project FP7 – 2010-2013a
http://www.sapere-project.eu
http://apice.unibo.it/xwiki/bin/view/SAPERE/
under the hat of the Proactive Initiative AWARENESS
http://www.aware-project.eu/
a
This work has been supported by the EU-FP7-FET Proactive project
SAPERE – Self-aware Pervasive Service Ecosystems, under contract no.256873
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 17 / 36
18. SAPERE
Self-aware Pervasive Service Ecosystems II
SAPERE model
based on chemical coordination for pervasive computing
[Viroli et al., 2012, Zambonelli et al., 2011]
LSA (Live Semantic Annotation), as chemical tuples representing
individuals, components, services in pervasive scenarios, and
triggering eco-laws governing self-organisation of pervasive services
LSA are reified into LSA-spaces, representing contexts for distributed
nodes
LSA-bonding allows an LSA to link to other LSA in the same space,
and SAPERE entities to inspect the state of their peers and act
accordingly
system rules – the eco-laws – are instead in charge of managing the
global behaviour of the whole system, by manipulating – deleting,
updating, moving, bonding – LSA in the system, in the way of
chemical reactions
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 18 / 36
19. Model-checking: A Possible Approach
Outline
1 Motivations
2 Steps
3 Background
4 SAPERE
5 Model-checking: A Possible Approach
6 Data Mining: A Possible Approach
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 19 / 36
20. Model-checking: A Possible Approach
MABLE I
Tools for model checking
model checking is an undoubtedly effective analysis, less often efficient
which language is used to model the system as well as which tools are
used for the analysis play a very important role
the main issue is the ability to check global properties of MAS, as well
as properties of the internal state of the individual entities
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 20 / 36
21. Model-checking: A Possible Approach
MABLE II
MABLE
MABLE [Wooldridge et al., 2002] is a C-like imperative language that
allows the definition of MAS and the verification of system properties
expressed in LTL
the system modelled using MABLE is translated into a Promela
specification that can be analysed using compatible tools such as
SPIN
currently developed a translation system to Java
MABLE has two unique features compared to other similar languages:
claims and semantic specifications
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 21 / 36
22. Model-checking: A Possible Approach
MABLE III
Claims
claims allow the definition of properties to be verified during model
checking
these properties are not limited to individual entities, but they affect
the system in its entirety
Semantic specifications
a semantic specification allows the definition of pre- and
post-conditions for a given communication in the
MAS[Huget and Wooldridge, 2004]
similarly to SAPERE eco-laws
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 22 / 36
23. Model-checking: A Possible Approach
MABLE IV
Assumptions on internal agent states
another feature of MABLE is the capability of an agent to make
assumptions on the internal state of other agents—on their beliefs,
for instance
assumptions can also be expressed inside claims, which is particularly
useful in MAS contexts, where entities are essentially blind to the
internal state of others
this could be used to model the SAPERE mechanism of LSA-bonding
SAPERE, which would change an external piece of information (LSA)
into a internal property of the agent (belief)
a bond between LSA would simply be represented as a a connection
of the beliefs of the agents
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 23 / 36
24. Model-checking: A Possible Approach
MABLE V
Crowd steering in SAPERE
two possible ways
modelling everything represented by a LSA as a MABLE agent
modelling only nodes and user handheld devices as agents, and other LSA as
internal beliefs of the node agent
in the former approach, fields of the LSA would become beliefs of the corresponding agent,
and bonds between LSA would become beliefs on the internal state of the coupled agents
in the latter, a new LSA corresponds to a communication act towards the node agent,
followed by a change of its internal beliefs
in both ways, the eco-laws would be modelled as MABLE communication acts, with pre-
and post-conditions
claims could be used to check whether the crowd level in the nodes, specified by an
internal belief of the agent corresponding to the node, evolves in a satisfactory way
a SAPERE model could be represented in MABLE, then translated into a Promela
specificationa , and ideally analysed using SPINb —or, any other model checker supporting
Promela specifications
a
Promela verification language:
http://www.dai-arc.polito.it/dai-arc/manual/tools/jcat/main/node168.html
b
The SPIN model checker: http://spinroot.com/spin/whatispin.html
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 24 / 36
25. Data Mining: A Possible Approach
Outline
1 Motivations
2 Steps
3 Background
4 SAPERE
5 Model-checking: A Possible Approach
6 Data Mining: A Possible Approach
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 25 / 36
26. Data Mining: A Possible Approach
Data Mining I
Issues
putting together the techniques presented in [Kaminka et al., 2003]
with the results described in [Lattner et al., 2006], one might be able
to face the issue of how the parameters of a simulation could be
tuned up, and how to set the relation between the values of these
parameters and the corresponding emergent behaviour
the key-issue in data mining are primarily the identification of the
values that define the emergent behaviour, and the creation of a
significant and adequately representative data-set
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 26 / 36
27. Data Mining: A Possible Approach
Data Mining II
Goals
1 the correlation between parameter values and emergent behaviour
2 possibly counter-intuitive correlations between some parameters and
some aspects of the emergent behaviour could be pointed out
3 identification of the ideal assignments of values, to either enforce
desired behaviours, or discourage undesired ones
4 identification of correlations between micro- and macro-events, as well
as between macro-events themselves, such as the identification of
recurrent patterns of the emergent behaviour
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 27 / 36
28. Data Mining: A Possible Approach
Data Mining III
Crowd Steering
1 apart from the many algorithms that could be useful—partitional
(K-means), if the number of clusters is known beforehand,
hierarchical (single linkage), if it is not, bayesian, density based
(DBSCAN), neural networks, or SVM (Support Vector Machine)
2 the first step would be to identify the parameters and values to be
analysed—for instance, the crowd factor that penalises the calculation
of distances, the rate of the eco-laws, and the effect of changing
those values on the crowd levels in every node
3 to build a meaningful training set, a variety of simulation should be
executed, varying both the values of the parameters and the topology
of the museum, so to prevent pathological cases to affect the analysis
4 Alchemist would be obviously the first choice here
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 28 / 36
29. Conclusion
Position paper, of course. . .
we just started facing the issue of analysing and predicting the
behaviour of complex pervasive, self-organising systems like SAPERE
coordinated systems
we shortly survey the literature on the general subject along with the
available tools, then we select two promising techniques and sketch
their application to the analysis and prediction of the behaviour of
SAPERE pervasive ecosystems
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 29 / 36
30. Conclusion
Future works
actually experimenting the two approaches here devised upon selected
SAPERE applications – such as crowd steering and pervasive display
applications
trying to integrate different techniques into a single articulated yet
coherent approach capable of providing reliable prediction of the most
meaningful and useful features of a SAPERE system
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 30 / 36
31. Bibliography
Bibliography I
Behrens, T. M. and Dix, J. (2007).
Model checking multi-agent systems with logic based Petri nets.
Annals of Mathemathics and Artificial Intelligence, 51(2-4):81–121.
From the issue entitled “Special Issue in honor of Professor Jack Minker’s 80th birthday”.
Bordini, R. H., Fisher, M., Pardavila, C., and Wooldridge, M. J. (2003).
Model checking AgentSpeak.
In Rosenschein, J. S., Wooldridge, M. J., Sandholm, T., and Yokoo, M., editors, 2nd
International Joint Conference on Autonomous Agents and Multi-Agents Systems
(AAMAS 2003), pages 409–416, New York, NY, USA. ACM Press.
Bordini, R. H., Fisher, M., Visser, W., and Wooldridge, M. J. (2004).
Verifiable multi-agent programs.
In Dastani, M. M., Dix, J., and El Fallah-Seghrouchni, A., editors, Programming
Multi-agent Systems, volume 3067 of LNCS, pages 72–89. Springer.
1st International Workshop (PROMAS 2003), Melbourne, Australia, 15 July 2003.
Selected Revised and Invited Papers.
Bordini, R. H., Fisher, M., Visser, W., and Wooldridge, M. J. (2006).
Verifying multi-agent programs by model checking.
Autonomous Agents and Multi-Agent Systems, 12(2):239–256.
EUMAS’04 Special Issue.
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 31 / 36
32. Bibliography
Bibliography II
Dixon, C., Winfield, A., and Fisher, M. (2011).
Towards temporal verification of emergent behaviours in swarm robotic systems.
In Groß, R., Alboul, L., Melhuish, C., Witkowski, M., Prescott, T. J., and Penders, J.,
editors, Towards Autonomous Robotic Systems, volume 6856 of LNCS, pages 336–347.
Springer.
12th Annual Conference (TAROS 2011), Sheffield, UK, 31 August – 2 September 2011.
Proceedings.
Giordano, L., Martelli, A., and Schwind, C. (2004).
Verifying communicating agents by model checking in a temporal action logic.
In Alferes, J. J. and Leite, J., editors, Logics in Artificial Intelligence, volume 3229 of
LNCS, pages 57–69. Springer.
Huget, M.-P. and Wooldridge, M. J. (2004).
Model checking for ACL compliance verification.
In Dignum, F., editor, Advances in Agent Communication, volume 2922, pages 75–90.
Springer.
International Workshop on Agent Communication Languages (ACL 2003), Melbourne,
Australia, 14 July 2003. Revised and Invited Papers.
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 32 / 36
33. Bibliography
Bibliography III
Kaminka, G. A., Fidanboylu, M., Chang, A., and Veloso, M. M. (2003).
Learning the sequential coordinated behavior of teams from observations.
In Kaminka, G. A., Lima, P. U., and Rojas, R., editors, RoboCup 2002: Robot Soccer
World Cup VI, volume 2752 of LNCS, pages 111–125. Springer.
Konur, S., Dixon, C., and Fisher, M. (2010).
Formal verification of probabilistic swarm behaviours.
In Dorigo, M., Birattari, M., Di Caro, G. A., Doursat, R., Engelbrecht, A. P., Floreano, D.,
Gambardella, L. M., Groß, R., Sahin, E., Sayama, H., and St¨tzle, T., editors, Swarm
u
Intelligence, volume 6234 of LNCS, pages 440–447. Springer.
7th International Conference (ANTS 2010), Brussels, Belgium, 8-10 September 2010.
Proceedings.
Lattner, A. D., Miene, A., Visser, U., and Herzog, O. (2006).
Sequential pattern mining for situation and behaviour prediction in simulated robotic
soccer.
In Bredenfeld, A., Jacoff, A., Noda, I., and Takahashi, Y., editors, RoboCup 2005: Robot
Soccer World Cup IX, volume 4020 of LNCS, pages 118–129. Springer.
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 33 / 36
34. Bibliography
Bibliography IV
Mnif, M., Richter, U., Branke, J., Schmeck, H., and M¨ller-Schloer, C. (2007).
u
Measurement and control of self-organised behaviour in robot swarms.
In Lukowicz, P., Thiele, L., and Tr¨ster, G., editors, Architecture of Computing Systems
o
—- ARCS 2007, volume 4415 of LNCS, pages 209–223. Springer.
20th International Conference, Zurich, Switzerland, 12-15 March 2007. Proceedings.
Pokorny, L. R. and Ramakrishnan, C. R. (2006).
Modeling and verification of distributed autonomous agents using logic programming.
In Leite, J. A., Omicini, A., Torroni, P., and Yolum, P., editors, Declarative Agent
Languages and Technologies II, volume 3476 of LNAI, pages 313–319. Springer.
2nd International Workshop (DALT 2004), New York City, NY, USA, 19 July 2004.
Revised Selected Papers.
Viroli, M., Pianini, D., Montagna, S., and Stevenson, G. (2012).
Pervasive ecosystems: a coordination model based on semantic chemistry.
In Ossowski, S., Lecca, P., Hung, C.-C., and Hong, J., editors, 27th Annual ACM
Symposium on Applied Computing (SAC 2012), Riva del Garda, TN, Italy. ACM.
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 34 / 36
35. Bibliography
Bibliography V
Wooldridge, M. J., Fisher, M., Huget, M.-P., and Parsons, S. (2002).
Model checking multi-agent systems with MABLE.
In Castelfranchi, C. and Johnson, W. L., editors, 1st International Joint Conference on
Autonomous Agents and Multiagent Systems (AAMAS 2002), volume 2, pages 952–959,
New York, NY, USA. ACM.
Zambonelli, F., Castelli, G., Ferrari, L., Mamei, M., Rosi, A., Di Marzo, G., Risoldi, M.,
Tchao, A.-E., Dobson, S., Stevenson, G., Ye, Y., Nardini, E., Omicini, A., Montagna, S.,
Viroli, M., Ferscha, A., Maschek, S., and Wally, B. (2011).
Self-aware pervasive service ecosystems.
Procedia Computer Science, 7:197–199.
Proceedings of the 2nd European Future Technologies Conference and Exhibition 2011
(FET 11).
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 35 / 36
36. Towards the Analysis & Prediction of Complex System
Behaviour in SAPERE
Marco Alberti & Andrea Omicini
marco.alberti3@studio.unibo.it, andrea.omicini@unibo.it
Alma Mater Studiorum—Universit` di Bologna
a
ASENSIS 2012
1st International Workshop @ SASO 2012,
Lyon, France, 10th of September 2012
Alberti & Omicini (Universit` di Bologna)
a Towards SAPERE Analysis & Prediction ASENSIS 2012, 10/9/2012 36 / 36