Presentation is loading. Please wait.

Presentation is loading. Please wait.

Spring-2007 Vagan Terziyan

Similar presentations

Presentation on theme: "Spring-2007 Vagan Terziyan"— Presentation transcript:

1 Spring-2007 Vagan Terziyan
TIES-423 (TLI363) – Agent Technologies in Mobile Environment former name: TLI371 – Distributed Artificial Intelligence in Mobile Environment Course Introduction Vagan Terziyan Department of Mathematical Information Technology University of Jyvaskyla ;

2 Contents Course Introduction Lectures and Links Course Assignment
Course Exercise

3 Practical Information
12 Lectures (2 x 45 minutes each, in English) during period 12 March - 24 April according to schedule: 8 lectures by Vagan Terziyan – theory; 4 lectures by Artem Katasonov – theory and practice; 4 Laboratory works in computer class (2 x 45 minutes each, in English) during period 7 May - 15 May according to schedule, by Artem Katasonov; Slides for lectures: available online; Assignment. Based on the theoretical part of the course. Make PowerPoint presentation based on a research paper); Group Exercise. Based on the practical part of the course and related to design of a multi-agent system with SmartResource Platform (a tool on the top of JADE); Exercise and assignment should be available for review until 31 May (24:00); Exam: There will be no exam. Course grade will be given based on the exercise and assignment quality.

4 Lectures Topics and Schedule (1)
12 March 2007 – Course Introduction (today) Lecture 1 - ”Agent Technologies in Mobile Environment: Course Introduction” 13 March 2007 – Overview of Intelligent Agents Lecture 2 - ”What is an Intelligent Agent ?” 19 March 2007 – Overview of (Multi)Agent Technologies - I Lecture 3 - ”Agent Technologies - I” 20 March 2007 – Overview of (Multi)Agent Technologies - II Lecture 4 - ”Agent Technologies - II” 26 March 2007 – Agent Intelligence – I Lecture 5 - ” Agent Logic, Reasoning and Planning” 27 March 2007 – Agent Intelligence – II Lecture 6 - ” Agent Learning and Knowledge Discovery” 2 April 2007 – Industrial Applications of Agent Technology - I Lecture 7 - ”SmartResource: Agent-Based Self-Managed Web Resources - I” 3 April 2007 – Industrial Applications of Agent Technology - II Lecture 8 - ”SmartResource: Agent-Based Self-Managed Web Resources - II” Ag. C134.1 Ag. Auditorio 2 Ag. C233.1 Monday lectures: 12:15 – 13:55; Break: 13:00 – 13:10; Place: Agora Alfa Tuesday lectures: 10:15 – 11:55; Break: 11:00 – 11:10; Place: Agora Alfa

5 Lectures Topics and Schedule (2)
16 April – Agents as a Novel Software Engineering Paradigm Lecture 9 - ” Agent-Oriented Software Engineering” 17 April 2007 – Agent Platforms Lecture 10 - ”Agent Standards and Platforms” 23 April – Introduction to JADE Programming Lecture 11 - ”Introduction to JADE” 24 April – Development with SmartResource Platform Lecture 12 - ”SmartResource Platform” 7 May – Agent Design Lab - I Lab. work 1 - ”Getting started with JADE” 8 May – Agent Design Lab - II Lab. work 2 - ”Development for SmartResource I” 14 May – Agent Design Lab - III Lab. work 3 - ” Development for SmartResource II” 15 May – Agent Design Lab - IV Lab. work 4 - ” Development for SmartResource III” Place: Computer Class Monday lectures: 12:15 – 13:55; Break: 13:00 – 13:10; Place: Agora Alfa Tuesday lectures: 10:15 – 11:55; Break: 11:00 – 11:10; Place: Agora Alfa

6 Course Motivation Growing complexity of computer systems and networks used in industry  need for new approaches to manage and control them IBM vision: Autonomic computing – Self-Management (includes self-configuration, self-optimization, self-protection, self-healing) Ubiquitous computing, “Internet of Things”  huge numbers of heterogeneous devices are interconnected “nightmare of pervasive computing” when almost impossible to centrally manage the complexity of interactions, neither even to anticipate and design it. We believe that self-manageability of a complex system requires its components to be autonomous themselves, i.e. be realised as agents. Agent-based approach to SE is also considered to be facilitating the design of complex systems

7 INTEL: Proactive Computing Concept (1)
Intel Research initiated work on Proactive Computing (beginning 2001) - working towards environments in which networked computers proactively anticipate our needs and, sometimes, take action on our behalf. Intel identified three steps that are essential to making proactive computing a reality: The first is getting physical — connecting billions of computing devices directly to the physical world around them so that human beings are no longer their principal I/O devices. The next step is getting real — having computers running in real time or even ahead of real time, anticipating human needs rather than simply responding to them; The third step is getting out — extending the role of computers from the office and home into the world around us and into new application domains.

8 INTEL: Proactive Computing Concept (2)
“Intel Research is exploring computing futures that overlap autonomic computing but also explore new application domains that require principles we call proactive computing, enabling the transition from today’s interactive systems to proactive environments that anticipate our needs and act on our behalf.” (R. Want, T. Pering, D. Tennenhouse, Comparing Autonomic and Proactive Computing, IBM Systems Journal, Vol 42, No 1, 2003) Proactive system design is guided by seven underlying principles: connecting with the physical world, deep networking, macro-processing, dealing with uncertainty, anticipation, closing the control loop, making systems personal.

9 IBM: Autonomic Computing (1)
The computing domain is now a vast and diverse matrix of complex software, hardware and services. By 2020 we expect billions of devices and trillions of software processes, with a lot of data. And it's not just a matter of numbers. It's the complexity of these systems and the way they work together that is creating a shortage of skilled IT workers to manage all of the systems. It's a problem that's not going away, but will grow exponentially, just as our dependence on technology has. Autonomic Computing is about how to enable computing systems to operate in a fully autonomous manner. No administration, just simple high-level policy statements. Autonomic Computing is an approach to self-managed computing systems with a minimum of human interference. The term derives from the body's autonomic nervous system, which controls key functions without conscious awareness or involvement.

10 IBM: Autonomic Computing (2)

11 IBM: Service-Oriented Architecture (1)
Message from the Vice President, IBM Asset and Integration Technology, Software Group “As we regard the advances that have moved us into the 21st century, we observe that information technology (IT) seems to repurpose itself almost every year. Like the invention of transistors … the new service-oriented thinking and its application to IT known as service-oriented architecture (SOA) distinguishes itself as a paradigm change. Seen in the context of an entirely new service-oriented “business ecosystem,” SOA could be one of the most significant technological advances, enabling the IBM corporate strategy of business on demand...” “Business processes must be decomposed, services must be created, and the supporting machinery must be implemented, so that the business ecosystem can run effectively, efficiently, and manageably.” “IBM has found that businesses which made the transition to service-oriented enterprises have shown significant savings in maintenance, personnel, and software and hardware costs. This transition starts with the use of the Component Business Model (CBM) … and continues with the application of Service Oriented Modeling and Architecture (SOMA)...”

12 IBM: Service-Oriented Architecture (2)
In the current business environment in which companies are under increasing pressure not only to increase revenue but also to respond quickly to changing market conditions, companies will be successful only if they transform themselves and become on demand businesses. Needed transformation changes include componentization and service-orientation. Componentization enables a business to operate in a value net, a network of partnerships with customers and suppliers supported by real-time information flows and information technology systems. Service-orientation is needed to achieve seamless integration of business components. Recent IBM activities and experiences in this area prove high business value for these challenges. L. Cherbakov, G. Galambos, R. Harishankar, S. Kalyana, and G. Rackham, Impact of service orientation at the business level, In: Service-Oriented Architecture, IBM Systems Journal ,   Volume 44, Number 4, December 2005.

13 Norwegian University of Science and Technology, Trondheim
TAPAS The “Theatre” metaphor Theatre: A metaphor for concepts and functionality definition. Repertoire: The set of Plays that may be performed at the theatre. Play: Defines a set of logically related functionality. Director role figure: The manager of plays, and supervisor for application role figures, constituted by an actor . Application role figures : The performers of plays. Constituted by actors playing roles. Capability: A unique set of properties of an actor at the stage where he is playing. Role session: A dialogue between two role figures. Actors Manuscript: The assigned behavior, i.e. the defined role of a role figure, constituted by an an actor. Norwegian University of Science and Technology, Trondheim

14 Google: Excellent content and context provider for Web applications
Google Maps, Google Earth, Wikimapia, GMail, Blogger, etc.

15 Two alternative trends of Web development
Machines, devices, software, etc Human Communities Semantic Web Web 2.0 Facilitates Human-to-Human interaction Wikis Metadata Blogs Ontologies Facilitates Machine-to-Machine interaction SW Services Mashups Agents Web-Services Web EAI Portals Community Portals

16 What is Wiki Wiki is the simplest online database that could possibly work. Wiki is a piece of server software that allows users to freely create and edit Web page content using any Web browser. Wiki supports hyperlinks and has a simple text syntax for creating new pages and crosslinks between internal pages on the fly. Wiki is unique among other group communication mechanisms because it allows editing the organization of content in addition to the content itself. Wiki encourages democratic use of the Web by promoting content composition by non-technical users.

17 Collaborative editing window
Sample of Wiki Web page Collaborative editing window

18 Wikipedia

19 Web 2.0 Community Portal

20 Motivation for Semantic Web

21 Semantic Web: New “Users”
applications agents

22 Web resources / services / DBs / etc.
Semantic Web: Resource Integration Semantic annotation Shared ontology Web resources / services / DBs / etc.

23 This is just a small part of Semantic Web concern !!!
Semantic Web: which resources to annotate ? This is just a small part of Semantic Web concern !!! Technological and business processes External world resources Web resources / services / DBs / etc. Semantic annotation Shared ontology Multimedia resources Web users (profiles, preferences) Smart machines, devices, homes, etc. Web agents / applications / software components Web access devices and communication networks

24 GUN Concept GUN – Global Understanding eNvironment GUN =
Global Environment + Global Understanding Proactive Self-Managed Semantic Web of Things = (we believe) = “Killer Application” for Semantic Web Technology

25 GUN and Ubiquitous Society
GUN can be considered as a kind of Ubiquitous Eco-System for Ubiquitous Society – the world in which people and other intelligent entities (ubiquitous devices, agents, etc) “live” together and have equal opportunities (specified by policies) in mutual understanding, mutual service provisioning and mutual usability. Human-to-Human Human-to-Machine Machine-to-Human Machine-to-Machine Agent-to-Agent

26 Core technologies for GUN
Distributed Artificial Intelligence Semantic Technology Interoperability, Automation and Integration Reusable semantic history blogs Reusable semantic behavior patterns and process descriptions Reusable coordination, design, integration and composition patterns Reusable decision-making patterns Reusable interface patterns Reusable security and privacy policies Proactivity Autonomic behavior Communication, coordination, negotiation, contracting Self-Configuration and Self-Management Learning based-on liveblog histories; Data Mining and knowledge discovery; Dynamic integration; Diagnostics and prediction; Model exchange and sharing

27 GUN-GERI-UBIWARE-SmartResource ?
GUN GERI UBIWARE SmartResource GUN (Global Understanding Environment) – Proactive Self-Managed Semantic Web of Things - general concept and final destination GERI (Global Enterprise Resource Integration) – GUN subset related to industrial domains UBIWARE – middleware for GERI SmartResource – semantic technology, pilot tools and standards for UBIWARE

28 SmartResource in the IOG Web Site

29 One of Smart Resource Scenarios
“Knowledge Transfer from Expert to Service” Agent plays roles: Scene 1: “diagnostic expert”; Scene 2: “no play”; Scene 3: “no play” “Expert” Labelled data Agent plays roles: Scene 1: “no play”; Scene 2: “student”; Scene 3: “diagnostic expert” Watching and querying diagnostic data Querying diagnostic results “Device” Labelled data Labelled data “Service” Labelled data History data Querying data for learning Learning sample and Querying diagnostic results Agent plays roles: Scene 1: “patient”; Scene 2: “teacher”; Scene 3: “patient” Diagnostic model

30 Agent-driven EAI (1) Data Warehouse USERS UBIWARE operator field crew
expert consumers owner manager administration USERS UBIWARE Production External Applications Services Maintenance Intelligence Enterprise portal Automation Data Warehouse

31 AI tools (Knowledge Discovery) Sensors and alarm detectors
Agent-driven EAI (2) AI tools (Knowledge Discovery) Sensors and alarm detectors Software and services Maintenance workers Operators Experts UBIWARE Resource info Other users UBIWARE Industrial Resource Data Warehouse

32 Agents in mobile environment

33 Agent-driven EAI in mobile environment
customers manager administration Call center field crew Expert/specialist UBIWARE Data Warehouse Intelligence GPS

34 Agent-driven integration in mobile environment
3G WWAN Operating on 3G WWAN Zone 1 Zone 2 Zone 3 Plug into power jack Wakeup Wi-Fi Continue over Wi-Fi Wakeup Wi-Fi Zone 4 Zone 5 Zone 6 Airport Radio State 3G WWAN Wi-Fi WiMAX GPS Radio State 3G WWAN Wi-Fi WiMAX GPS Radio State 3G WWAN Wi-Fi WiMAX GPS Radio State 3G WWAN Wi-Fi WiMAX GPS Radio State 3G WWAN Wi-Fi WiMAX GPS Radio State 3G WWAN Wi-Fi WiMAX GPS Radio State 3G WWAN Wi-Fi WiMAX GPS Wi-Fi Link Going Down. Connect to Wi-Fi Home Continue session on 3G WWAN Continue session on Wi-Fi Battery level low Shutdown WiMAX Switch to 3G WWAN Operator initiated switch to WiMAX Continue session on WiMAX Shutdown Wi-Fi Zone 7 WiMAX Zone 8 WiMAX Zone 9 IEEE for Network Discovery IEEE , SIP, VCC, IMS, for Network Selection and Service Continuity across multiple radios (3G WWAN  Wi-Fi  WiMAX) 802.21, SIP, IMS for Service Continuity (Wi-Fi  WiMAX) VCC, SIP, IMS for Call Continuity (3G WWAN  Wi-Fi)

35 Agent-driven peer-to-peer environments
JADE-LEAP Agent Platform is extension to JADE (special container within JADE) Target devices Java MIDP-capable phones PDA devices Smallest available platform in terms of footprint size Proprietary device-initiated and socket based communication channel with main container Developed within LEAP project Open-source Mikko Laukkanen

36 Agent-Driven EAI (Human-Centric)
2 Human as UBIWARE service provider Sensing Online Monitoring Testing Diagnostics Treatment UBIWARE Human as UBIWARE Resource (i.e. service consumer) Human as UBIWARE user (utilizing integrated data and knowledge) Human as UBIWARE administrator 4 Data Warehouse UBIWARE 3 1

37 Word-Wide Correlated Activities
Semantic Web Agentcities is a global, collaborative effort to construct an open network of on-line systems hosting diverse agent based services. Semantic Web is an extension of the current web in which information is given well-defined meaning, better enabling computers and people to work in cooperation Agentcities Grid Computing Wide-area distributed computing, or "grid” technologies, provide the foundation to a number of large-scale efforts utilizing the global Internet to build distributed computing and communications infrastructures. FIPA FIPA is a non-profit organisation aimed at producing standards for the interoperation of heterogeneous software agents. Web Services WWW is more and more used for application to application communication. The programmatic interfaces made available are referred to as Web services. The goal of the Web Services Activity is to develop a set of technologies in order to bring Web services to their full potential

38 Java programming, AI basics
Package of courses Java programming, AI basics Spring Fall Design of distributed, self-descriptive, autonomous, proactive, self-managed, interoperable, intelligent systems, applications and services

39 ATME Course: Lectures

40 Lecture 1: This Lecture - ATME Introduction

41 Lecture 2: What is an Intelligent Agent ?

42 Lectures 3-4: Agent Technologies (Mobility, Communication, Coordination, Negotiation)

43 Lectures 5-6: Agent Intelligence (Internal Logic, Reasoning, Planning, Learning, Knowledge Discovery)

44 Lectures 7-8: Industrial Applications of Agent Technology: SmartResource - Agent-Based Self-Managed Web Resources

45 Lecture 9: Agents as a Novel Software Engineering Paradigm
Benefits Agent platforms and agent programming languages (APL) Potential effect on problem analysis and requirements processes This and following lectures are by Artem Katasonov

46 Lecture 10: Agent Platforms
FIPA (IEEE) architecture Existing platforms: JADE Cougaar AgentFactory 3APL Jason (AgentSpeak APL) SmartResource Platform

47 Lecture 11: Introduction to JADE
Architecture System agents and their GUIs Main classes (Agent, Behaviour) and their abilities see also:

48 Lecture 12: SmartResource Platform
Architecture Script language (semantic APL) Developing Reusable Atomic Behaviors (RABs)

49 ATME Course: Assignment

50 Assignment in brief Students are expected to select one of below recommended papers (or any other relevant research paper from the Web) and make PowerPoint presentation based on that paper. The presentation should provide evidence that a student has got the main ideas of the paper, is able to provide his personal additional conclusions and critics to the approaches used.

51 Evaluation criteria for the assignment
Content and Completeness; Clearness and Simplicity; Discovered Connections to ATME Course Material; Originality, Personal Conclusions and Critics; Design Quality.

52 Format, Submission and Deadlines
Format: PowerPoint .ppt , name of file is student’s family name; Presentation should contain all references to the materials used, including the original paper; Deadline - 31 May 2007 (24:00); Files with presentations should be sent by to Vagan Terziyan and Notification of evaluation - until 10 June.

53 Papers for Course Assignment (1)

54 Papers for Course Assignment (2)

55 ATME Course: Group Exercise

56 Group Exercise in brief
In small groups of 2-4 people Based on the practical part of the course and related to design of a multi-agent system with SmartResource Platform. At least some members of the group should have some experience in JAVA programming (for developing RABs). Since a major part of development work under SmartResource Platform is done through high-level scripting in semantic APL, students without experience in JAVA can participate as well, taking these tasks. Deadline - 31 May 2007 (24:00); Source files and minimal documentation should be sent by to Artem Katasonov

57 Information about Related Course
Agent Technologies in the Semantic Web ; by Vadim Ermolayev; recommended as additional reading.

58 Additional reading (1): Agent Reasoning with Uncertainty: Introduction to Bayesian Networks

59 Additional Reading (2): Personalization in Mobile Environment

60 Additional slides (old content)

61 Distributed AI Applications
Application Area Web Content Management Emerging Application Personalization Distributed transactions management Agent technologies Solutions Profile / Location management Data mining Beliefs management Knowledge metamodeling Filtering

62 Some Professions around Semantic Web
Content creators AI Professionals Content Logic, Proof and Trust Mobile Computing Professionals Ontologies Web designers Agents Annotations Ontology engineers Software engineers

63 Lecture n: Agents for Personalizing Web Resources: Web Content Personalization Overview

64 Lecture nn: Collaborative Filtering

65 Lecture nnn: Similarity Evaluation Techniques for Filtering

66 Lecture nn: Agent-based Knowledge Discovery: Dynamic Integration of Virtual Predictors

67 JADE (Java Agent DEvelopment Platform)

68 Agent Standards: FIPA Agent Framework

69 ATME Course: Exercise

70 Task for the Exercise (according to A. Raja) (1)
Consider the home of the future where there are software agents in a mobile environment that are helping to manage the running of a house. There will be: (1) Personal assistant agents that will know of your preferences of temperature, humidity, light, sound, etc., and who you want to interact with; (2) There will be agents that can measure appropriate environmental conditions with specific devices; (3) There will be agents that effect appropriate environmental conditions with specific devices; (4) There will be agents that control expenses for the use of appropriate devices; (5) There will be agents that manage the telephone communications; (6) There will be agents that manage security issues such as fire, earthquake, flood protection, etc.

71 Task for the Exercise (according to A. Raja) (2)
Assume that the agents are heterogenous (i.e. have not be generated by one designer), for example when you get a new device it will come with an agent; for instance, the heating measurement agent may not come from the same company as the air-conditioning agent. Think about the possibility of having these agents work together. What are the capabilities of the agents, what type of cooperation needs to occur among them, are there needs for the agents to negotiate, are there situations where local objectives are at odds with global objectives such as minimizing electrical usage? What type of information needs to be exchanged among the agents?

72 Task for the Exercise (according to A. Raja) (3)
How would you organize the agents – would you have a hierarchy of agents in terms of their control responsibilities? How would you allow agents to integrate new agents into the system, for instance, when you buy a new device. What are the specific characteristics required by a language in order that these agents can share information? If there are no dedicated resources for each agent, but rather a pool of resources that can be used by agents, what new issues does this introduce? Do agents need to reason about the intentions of other agents?

73 Task for the Exercise (according to A. Raja) (4)
In answering these and related issues that you may consider, please be concrete with specific and numerous examples/scenarios. You should first start out the effort by detailing the collection of agents that you see in the house of the future, what their responsibilities are, and their patterns of interaction with other agents. Including figures, it should be at least 5 pages long.

74 Format, Submission and Deadlines
Format: MS Word doc. (winzip encoding allowed), name of file is student’s family name; Presentation should contain all references to the materials used; Deadline - 20 October 2004 (24:00); Files with presentations should be sent by to Vagan Terziyan Notification of evaluation - until 29 October.

Download ppt "Spring-2007 Vagan Terziyan"

Similar presentations

Ads by Google