awareness inaugural meeting amsterdam 2010
DESCRIPTION
Presentation Slides from the Awareness Inaugural Meeting Amsterdam 2010. Awareness is a Future and Emerging Technologies Proactive Initiative funded by the European Commission under FP7TRANSCRIPT
Inaugural Meeting for AWARENESS partners
14-15 December 2010Amsterdam
FP7: FET Proactive Intiative: Self-Awareness in Autonomic Systems (AWARENESS)
Monday, 3 January 2011
Introductions
Monday, 3 January 2011
Meeting objectives
Get to know each other betterUnderstand the Awareness projects betterAppreciate areas of commonality and where we can work closer together Understand AWARE’s activities and how you can participate, or how you can influence theseAppreciate where the AWARE CA can help your project Get to know each other better !
Monday, 3 January 2011
Agenda
Tuesday
LunchIntroduction of each project (approx 15 mins each)Overview of AWARE CACoffeeCommunity BuildingPublicity and DisseminationTraining activitiesEmerging Research ThemesOnline Features Magazine Dinner
Wednesday
FET conference May 2011Awareness summer school’11Training materials WorkshopsWebsiteNewsletters and shared infoResearch exchangesRoadmappingCommon DaysLunch
Monday, 3 January 2011
ASCENS: Autonomic Service-Component Ensembles
The ASCENS approach will focus on service-component ensembles (SCEs), hierarchical ensembles built from service components (SCs), simpler SCEs and knowledge units (K) connected via highly dynamic infrastructure.
Partners:LMU MunichUniversità di PisaUniversità di FirenzeFraunhofer GesellschaftVERIMAG LaboratoryUniversità di Modena e Reggio EmiliaLero - University of LimerickUniversite Libre de BruxellesEPF LausanneVolkswagen AGZimory GmbHISTI (Third Party)
Prof. Dr. Martin Wirsing (Coordinator)Universität München, Institut für Informatik
Monday, 3 January 2011
The EPiCS project aims at laying the foundation for engineering the novel class of proprioceptive computing systems. Proprioceptive computing systems collect and maintain information about their state and progress, which enables self-awareness by reasoning about their behaviour, and self-expression by effectively and autonomously adapt their behaviour to changing conditions.
Partners:
University of PaderbornImperial College LondonUniversity of OsloKlagenfurt UniversityUniversity of BirminghamEADS Innovation Works, MunichSwiss Federal Institute of Technology ZurichAustrian Institute of Technology GmbH, Vienna
EPiCS: Engineering Proprioception in Computing Systems
Prof. Dr. Marco Platzner (Coordinator)
University of Paderborn
Monday, 3 January 2011
The RECOGNITION project concerns new approaches for embedding self-awareness in ICT systems. This will be based on the cognitive processes that the human species exhibits for self-awareness, seeking to exploit the fact that humans are ultimately the fundamental basis for high performance autonomic processes.
Partners:
Cardiff UniversityItalian National Research CouncilUniversity of CambridgeNational and Kapodistrian University of AthensEurécomUniversity of Florence
RECOGNITION: Relevance and cognition for self-awareness in a
content-centric Internet
Prof. Roger M. Whitaker (Coordinator)
Cardiff University
Monday, 3 January 2011
The objective of SAPERE is the development of a highly-innovative theoretical and practical framework for the decentralized deployment and execution of self-aware and adaptive services for future and emerging pervasive network scenarios.
Partners:
Università di Modena e Reggio Emilia
Birkbeck College – University of London
The University Court of the University of St Andrews
Alma Mater Studiorum – Università di Bologna
Johannes Kepler Universitaet Linz
SAPERE: Self-Aware Pervasive Service Ecosystems
Prof. Franco Zambonelli (Coordinator)Università di Modena e Reggio Emilia
Monday, 3 January 2011
The main focus of SYMBRION is to investigate and develop novel principles of adaptation and evolution for symbiotic multi-robot organisms based on bio-inspired approaches and modern computing paradigms. Such robot organisms consist of super-large-scale swarms of robots, which can dock with each other and symbiotically share energy and computational resources within a single artificial-life-form.
Partners
Universitaet Stuttgart Universitaet Graz Vrije Universiteit Universitaet Karlsruhe Flanders Institute for Biotechnology University of the West of England, Bristol Eberhard Karls Universitaet Tuebingen University of York (Universite Libre de Bruxelles Institut National de Recherche en Informatique et Automatique
SYMBRION: Symbiotic Evolutionary Robot Organisms
(funded by PerAda)
Serge Kernbach (Coordinator)Universitaet Stuttgart
Monday, 3 January 2011
Overview of AWARE
AWARE Coordination Action in Self-Awareness in Autonomic Systems
Monday, 3 January 2011
Main Objectives: what we hope to achieve
To encourage greater cooperation and exchange between projects funded under the FET Proactive Initiative Awareness
To support researchers and encourage international collaboration
To improve visibility for the grand challenges and methodological approaches identified by this community
To support training activities including summer schools and exchange activities
To help expand a repository of knowledge for researchers (improved synchronisation of concepts, terminology, approaches, etc)
To organise a range of workshops and research consultation events
To promote the field more widely, generating interest with publishers, national science funding agencies, and within commercial environments.
Monday, 3 January 2011
Main Activities
Website: a constant presence and focal point for the community, providing information across a range of topics for a variety of users
A series of workshops for learning, information dissemination and knowledge transfer opportunities
Research exchanges to encourage greater interdisciplinary research
Summer schools to train the next generation of researchers and extending skills to the whole community
Research consultations and future roadmapping activities
Newsletters to provide regular updates of research news and events
An online Magazine promoting features on self-awareness research
The Awareness Book aimed at the general science public and considering wider socio-technical, socio-political and/or environmental impact
Documentaries, including website videos and other promotional vidoes to create a coherent thematic narrative of Awareness research aimed at engaging a wider audience.
Monday, 3 January 2011
Main Activities
All Coordination Actions aim to provide good cross-over between activities and the AWARE team works closely together
Community Building: Emma Hart
Publicity and Dissemination: Jeremy Pitt
Training: Gusz Eiben, Martijn Schut and Willem van Willigen
Emerging Research Themes: Giacomo Cabri
Supported by Jennifer Willies, Ingi Helgason and Callum Egan
AWARE Project Coordinator: Ben Paechter
Monday, 3 January 2011
Community Building
Why are we doing this?
Encourage research exchange and development
Support interdisciplinary research across national and international boundaries
Develop activities, promote events and disseminate useful materials
To encourage greater cooperation and exchange between people interested in self-awareness
Monday, 3 January 2011
Community Building Activities
Website as unified resource for Awareness community
Gather and publicise information about research and researchers, publications, surveys, articles; conference and workshop details; training materials
Regular AWARE newsletters and informative mailings
Workshops events in key research areas, ideally at major conferences preferred by Awareness projects
Annual exchange event involving all Awareness-funded projects aimed at cross-cutting themes and roadmapping objectives
Encourage greater international research cooperation by offering travel bursaries to researchers and inviting key international experts to Awareness events
Monday, 3 January 2011
Publicity and Dissemination
Why are we doing this?
Promote a common understanding of the science, technology and applications of self-aware systems across the range of Awareness projects.
Publicise and disseminate research results from the Awareness initiative, in an informative and accessible manner, through a number of conventional and innovative media.
Create lasting impact by producing tangible products whose utility to researchers and students will extend beyond the lifespan of the project
Monday, 3 January 2011
Publicity and Dissemination Activities
Website, promoting the public face of self-awareness in autonomic systems
Online Awareness magazine showcasing success stories and highlighting innovation and development in a popular science journalistic style
Awareness newsletter promoting ASCENS, EPiCS, RECOGNITION, SAPERE, SYMBRION and Awareness activities and events
Awareness book, an edited volume aimed at the general science public explaining the implications for science research
Awareness documentaries demonstrating project results, interviews with leading researchers to be disseminated via Awareness website, You Tube, and at workshops, science fairs, FET events
Monday, 3 January 2011
Training
Why are we doing this?
To promote training as a form of knowledge transfer to help influence European commercial competitiveness
To organise educational and training activities
To produce training materials useful to academia and industry
Monday, 3 January 2011
Training Activities
Three summer schools, particularly aimed at PhD students, post-docs or those new to the field
Production of training materials for an academic course (8-12 weeks) and an educated layman seminar (1-3 hours)
Collation of presentations from conferences to assist researchers
Build and maintain a web-based knowledge distribution system
Work with Awareness-funded projects to develop suitable training events
Monday, 3 January 2011
Emerging Research Themes
Why are we doing this?
Research pathfinding involving the Awareness community to determine strategic research directions
To identify potential for interdisciplinary cooperation across communities involved in Awareness-related research themes
To identify emerging research problems, key knowledge gaps and strategic developmental areas for problems related to self-aware and autonomic systems
Monday, 3 January 2011
Research Agenda Activities
Organising open web consultations to promote continuous dialogues including blog- and video interviews involving the Awareness projects
Organising consultation events bringing together researchers to identify key research issues (eg at FET11, main conferences)
Surveying and roadmapping within the Awareness community to provide an overview of research issues related to self-awareness in autonomic systems
Identifying potential synergies and complementarities within the research groups involved in the Awareness community, as well as with groups involved in other FET Proactive Initiatives
Monitoring relevant international research activities and initiatives
Monday, 3 January 2011
Specific Areas for Collaboration
Monday, 3 January 2011
Online Awareness magazine
Similar to PerAda magazine
60 feature articles highlighting innovation, like a journal
Promotion and explanation in 800 words; written in popular science style like New Scientist
Opportunity for wide audience and increase citations
Recommendations for good research stories
Monday, 3 January 2011
Awareness Keyword Cloud
Monday, 3 January 2011
Adaptation
• Adaption on multiple timescales• Organised adaptation• Adaptation to hostile situations• Adaptation to changing environments• Adaptation for robustness
Monday, 3 January 2011
Evolution/Emergence
Evolution• Evolution of new
collective behaviours• Open-ended evolution
Emergence• Emergent Systems• Emergent behaviours
Monday, 3 January 2011
Self-*
Self-properties• Self-expression• Self-optimisation• Self-organising networks• Self-organisation
Self-awareness• of state• about environment • of context• collective self-awareness• situation awareness
Monday, 3 January 2011
Awareness of me
• How do others see me ?• Look-* self-awareness• Is the environment aware of me ?
Monday, 3 January 2011
Learning/Behaviour
• Learning• Cognition• Filtering• Characteristics of behaviours• Opportunistic behaviour• Knowledge• Knowledge-intensive systems
Monday, 3 January 2011
Distribution and collectives
Distributed• Decentralised systems• Distributed artificial
intelligence• Robust distributed
systems• Distributed Control
Collectives• Collective intelligence• Global behaviour – local
decisions• Coordination
technologies• Collaborative decision
making
Monday, 3 January 2011
System Properties
Robust/Resilient• Fault tolerance• Robustness to sub-ideal
operation• Resilience
Others!• Relevant• Out of control• Homeostasis• Efficient• Autonomous
Monday, 3 January 2011
Socially Inspired
• Social media• Social cognition• Human cognition• Human in the loop• Augmented society• Social networking• Socio-technical combinatorics
Monday, 3 January 2011
Services/Systems
• Adaptive middleware• Architecture support for adaptivity• Self-joining services• Common services (middleware)• Service oriented architecture• Autonomic service components
Monday, 3 January 2011
Information and Modelling
Information/Recognition• Introspection about
norms and conventions• Utility of information• Intention recognition• Event recognition
Models• Meta-modelling of run-
time behaviour• Organisational models• Modelling the
environment• modelling inner state
Monday, 3 January 2011
Techniques/Systems
Techniques• Bio-inspired computing• Stream computing• Software-engineering• Pervasive computing• Social computing• E-mobility• Languages
• Measurement
Systems• Multi-agent systems• Ensembles• Self-governing
ensembles• Cloud computing• Sensor networks• Eco system• Robot swarms• Autonomous systems
Monday, 3 January 2011
Questions
• How does self-awareness relate to self-* ?• How are self-aware systems designed ?• What are meaningful applications of self-
awareness ?• How do we program such systems ?• How do we enable innovations ?
Monday, 3 January 2011
Open consultation and networking session on common related to self-awareness in autonomic systems
90 mins: decide best format
What are the key issues to address?
Proposal limited to 500 words and deadline is 15 Jan
FET11: 4-6 May in Budapest
From FET11 Call for Sessions:Should address a topic that is embryonic, multidisciplinary, transformative or foundational. Should aim to present state-of-the-art, develop broad visions and new concepts and identify resulting challenges for frontier research. Should feature a broad range of views, enabling different disciplines to come together and engage in a dialogue that creates a wider context.Highly interactive & unconventional session designs are welcome.
Selection criteria based on 1. Scientific and technological content
• novelty and interest of proposed topic, including possible creation of new area or transformation of existing area
• quality of proposed speakers• relevance to Future and Emerging Information Technologies• impact on science, technology or science policy• building of new collaborations, in particular across disciplines
2. Target group• key people/communities identified (e.g. diversity of actors)• level (not aimed too narrow/technical or too broad)• likely interest from addressed communities
3. Design and preparation• quality of session design• opportunity for interaction
Monday, 3 January 2011
Awareness Summer School 2011
Summer School 2011:
Early September : 5-6 days
Countryside, mountains or seaside : accommodation included
Anticipated numbers 25-35
Format: lectures and teamwork projects, practical examples, good social events, end-of-week presentations
Participants: PhD students, post-docs, yours?
Monday, 3 January 2011
Awareness Training Materials
Aim: to set up teaching repository on awareness
Focus now is downloadable slides/presentations
Focus later: might include text book
Templates for consistency wrt formatting and layout, but also wrt content, terminology, concepts etc
Input from summer school teaching materials, Awareness project, workshops
Weekend lock-in in a nice place!
Output to open courseware, tutorials, ITunesU, mobile apps (iOS, Android)
Bottom up process - input from projects
Visualisation and tag-cloud development
Monday, 3 January 2011
Organisation of workshops and support for workshops
Topics for workshops
What are the main conferences to aim for?
SASO 2011: Michigan Oct (SAPARE plus AWARENESS workshop)
ICAS 2011 Venice May
ICAC 2012 (intl conf autonomic computing)
SAKS 2011 Kiel, March
SAAES 2011 Algarve March
SEAMS 2011, Waikiki May
IROS San Francisco Sept
ACM-SAC
Workshop Planning
Monday, 3 January 2011
current website: www.aware-project.eu
More than the sum of our parts
Wordpress blog
Uses model of magazine/newspaper
Conversational (commenting system)
Main point of entrance to AwarenessTagged navigation: highly optimised for findability
Multimedia content for maximum publicity on activities and events
Repository for resources, CFPs, surveys
RSS/Twitter/Facebook
Monday, 3 January 2011
Optimisation of Awareness for search engines
• the key is to get the Information Architecture right
• reciprocal linking is hugely important, esp. for a ubiquitous term such as awareness
• keyword/phrase/theme density in web writing is equally important
• integrating the websites will help to push all sites up the search engine rankings
• by creating a highly visible research portal our research community will grow in numbers and across borders
• please contact me with any keywords/phrases/themes that are core to this research domain:[email protected]
• please link from your homepage to ours and link from your own web pages to each others and ours (thus, creating an AWARENESS network)
Monday, 3 January 2011
Awareness website
Interviews and short films on different subjects
Explaining project research to wider audiences
Awareness project documentaries
Monday, 3 January 2011
Newsletters, documentaries and shared information
Newsletters publicising project research, Awareness events, what is going on
Documentaries and website video clip
Let us help you with your project!
Let us capture your passions, your beliefs and your views on different subjects
Monday, 3 January 2011
Research exchanges
Six monthly simple application process
Aimed at multi-disciplinary collaboration between academics and/or industry
Contribution to travel/accommodation costs (need match funding, or in kind)
Short article for website to follow
Monday, 3 January 2011
Roadmapping consultations
Online blogging: how will this work?
Keyword recombination
What information researchers expect to find, and how will this help?
Online videos with experts’ opinions
Consultation events: what are the best formats and who to involve?
How best to represent the Living Document, how and who to shape it?
Monday, 3 January 2011
Awareness Common Days
First to be organised around Reviews next Oct?
Or at other suitable events?
Common subjects appropriate to most/all projects?
How to collate ideas and move forward?
Monday, 3 January 2011
What else can Awareness do for you?
Other ideas?
Over to you!
Contact us : www.aware-project.eu
Jennifer Willies: [email protected]
Callum Egan: [email protected]
Ingi Helgason: [email protected]
Monday, 3 January 2011
Future Emerging
Technologies
www.ascens-ist.eu
ASCENS Project
Martin Wirsing, Matthias Hölzl, Nora Koch
LMU Munich, Germany
AWARENESS meeting
Amsterdam, 14.12.2010
2 Seite
Partners
LMU Ludwig-Maximilians-Universität München
UNIPI Università di Pisa
UDF Università di Firenze together with
ISTI Istituto di Scienza e Tecnologie della Informazione “A. Faedo”
Fraunhofer Fraunhofer Gesellschaft (FIRST, Berlin)
VERIMAG Université Joseph Fourier Grenoble 1 (VERIMAG Lab.)
UNIMORE Università di Modena e Reggio Emilia Italy
ULB Université Libre de Bruxelles Belgium
EPFL Ecole Polytechnique Fédérale de Lausanne
VW Volkswagen AG
Zimory Zimory
UL University of Limerick (with University of Dublin)
2 www.ascens-ist.eu
3 Seite
Goal
3 www.ascens-ist.eu
Autonomous Service-Component
Ensembles
4 Seite
Autonomous Service-Component
Ensembles
Goal
4 www.ascens-ist.eu
5 Seite
Autonomous Service-Component
Ensembles
Heterogeneous
Massive number of nodes
Complex interactions or complex nodes
Open-ended, non-deterministic environment
Need to adapt to environment, new requirements
Goal
5 www.ascens-ist.eu
6 Seite
Building Ensembles
6 www.ascens-ist.eu
Open environments
Non-determinism
Changing requirements
Reliable
Predictable
Resilient
Fault tolerant
Complexity
8 Seite
Overview
SCEs
Knowledge
Self-awareness
Correctness
Foundational models
Case-studies
8 www.ascens-ist.eu
9 Seite
Overview
SCEs
Knowledge
Self-awareness
Correctness
Foundational models
Case-studies
9 www.ascens-ist.eu
Language
Engineering
Tool-integration
platform
10 Seite
Service-Component Ensembles
10 www.ascens-ist.eu
11 Seite
Research Topics
Language and Logics
Foundational Models
Correctness
Knowledge Representation
and Self-Awareness
Adaptation and Dynamic
Self-Expression
Tools and Tool
Integration
Engineering and
Best Practices
11 www.ascens-ist.eu
13 Seite
Case Studies
Self-Aware Robots
Science Cloud
e-Mobility
13 www.ascens-ist.eu
14 Seite
Thank You...
... for your attention!
14 www.ascens-ist.eu
Engineering Proprioception in Computing SystemsEngineering Proprioception in Computing SystemsEngineering Proprioception in Computing SystemsEngineering Proprioception in Computing Systems(EPiCS)(EPiCS)
—— Project Overview Project Overview ——
M Pl tMarco [email protected]
AWARE MeetingAWARE MeetingAmsterdamDecember 14-15, 2010
OutlineOutline•• Proprioceptive Computing Systems: PCSProprioceptive Computing Systems: PCS• EPiCS Consortium• EPiCS Consortium• Applications• Concepts and Foundations• Hardware/Software PlatformHardware/Software Platform• Networking and Middleware
Proprioceptive Computing Systems: PCSProprioceptive Computing Systems: PCS• PCS characteristics
– use proprioceptive sensors to monitor “one self” (concept from psychology, robotics/prosthetics, …, fiction)
– reason about their behaviour (self-awareness) – effectively and autonomously adapt their behaviour to
changing conditions (self-expression) proprioceptive sensors
• engineering PCS– transfer concepts of self-awareness/-expression
to computing and networking domains Learningto computing and networking domains– optimise performance and resource usage in
response to changing conditions – analyse limits for designing and operating
self-adaptive models of- the environment
lf
Self-awareness
y g g p gtechnological systems
• study suitability for different application domainsfinancial modeling on heterogeneous compute clusters
Feedback
- one self
Self-expression– financial modeling on heterogeneous compute clusters– person detection & tracking on distributed smart cameras – hypermusic on interactive mobile media systems
self-adaptive strategies
Self expression
2
Strategy
EPiCS ConsortiumEPiCS ConsortiumEPiCS: IP, 09/10-08/14, 8 partners from 5 countries, www.epics-project.eu
1. UPB (DE) University of Paderborn, Marco Platzner (coordinator)2 IMPERIAL (UK) I i l C ll L d W L k (WP3 l d)2. IMPERIAL (UK) Imperial College London, Wayne Luk (WP3 lead)3. UIO (NO) University of Oslo, Jim Torresen (WP5 lead)4. UNI-KLU (AT) Klagenfurt University, Bernhard Rinner5. UOBIRM (UK) University of Birmingham, Xin Yao (WP2 lead)6. EADS (DE) EADS Innovation Works Munich, Stephan Stilkerich7. ETHZ (CH) Swiss Federal Institute of Technology Zurich, Bernhard Plattner (WP4 lead)8. AIT (AT) Austrian Institute of Technology GmbH Vienna, Roman Pflugfelder
UNI-KLU, AIT
IMPERIAL
UIO UPB,
ETHZ
3
UIO
UOBIRM
,EADS
ApplicationsApplications• characteristics
– heterogeneous distributed systems, changing topologies, changing environments– high-performance and embedded computing– embedded into technical and non-technical contexts– classic design, operation, and management principles will fail
• heterogeneous compute cluster for financial modelling– accelerate workloads in data centres and cloud computing systems with latest
hardware technologies (GPU, FPGA) g ( , )– case studies: asset pricing and algorithmic trading
• distributed smart cameras for safety and security• distributed smart cameras for safety and security – real-time distributed embedded systems for computer vision using multiple cameras– case studies: person and object detection and tracking
• interactive mobile media system– active music allows the listener to control and adjust the music based on movements
4
– case study: hypermusic for joint music experience
App: Compute Cluster for Financial ModellingApp: Compute Cluster for Financial ModellingGigabit Ethernet Infiniband
System Memory
Heterogeneous Compute Node (HCN)
AMD PhenomQ d C
nVidia Tesla240 cores
Memory
Video Memory
Quad-Core
Xilinx FPGAcustom logic
Multi-Bank Memory
System I/O
PCI
HCN1
HCN0PCIe• challenges
– many code variantsdiff t Q S d d– different QoS demands
– dynamic scheduling to optimize speed or energyself optimization and self verification
5
HCNx– self-optimization and self-verification
App: Distributed Smart CamerasApp: Distributed Smart Cameras• challenges
– different appearance, occlusion – coordination of network resources – cooperation among cameras
cam node cam node cam node
TT Ethernet
middleware
6
application(comp. vision)mgt node
App: Interactive Mobile Media SystemApp: Interactive Mobile Media System• music
– pre-recorded– passive listeners
• hypermusic– programmed, partially pre-recorded– passive OR active listeners
– control (discrete)– start, stop, volume, fast fwd,
rewind, next/previous track
– control (continuous)– energy, expressivity, mood,
beat, pitch, timbre– social musical interaction
• challenge: extract and analyse motion information from sensors by machine learning to make new musical systems
Action Controller MappingSound Engine Sound
Instrument
7
Engine
Concepts and FoundationsConcepts and Foundations• research topics
– develop abstract models and clear problems for study – design algorithms for online learning in uncertain, dynamic, self-organisingdesign algorithms for online learning in uncertain, dynamic, self organising
environments, e.g. bio-inspired, consensus, and game-theoretic techniques; bandit solvers, ensemble learning
– develop mechanisms to ensure desirable global behaviourdevelop mechanisms to ensure desirable global behaviour– understand the effect of interacting nodes’ objectives,
strategies and behaviour on overall robustness, performance and quality of service
Learningp q y
– exploit self-awareness in order to learn to anticipate changes in the environment
– create software toolkit as a testbed and
self-adaptive models of- the environment- one self
Self-awareness
create software toolkit as a testbed and demonstration aid Feedback
o e se
Self-expression
self-adaptive strategies
8
Strategy
Hardware/Software PlatformHardware/Software Platform
CPU core(hardcore)
applications, quality of service requirements, Learning CPU core
(softcore)CPU core(softcore)
interconnect
(hardcore)
swthread
novel OS layer
Feedback
q ,system state
Self-awareness
(softcore)
swthread
novel OS layer
(softcore)
swthread
novel OS layer
interconnect
hwthread
novel OS layermonitoring core(proprioceptive
sensors)
Self-expression
Feedback
hwthread
novel OS layer
reconfigurable hardware core
sensors)
Strategy
thread assignment &migration, hardware reconfiguration, power & thermal management
reconfigurable hardware core
• research topicsdevelop architecture & operating system for autonomous heterogeneous multi core– develop architecture & operating system for autonomous heterogeneous multi-core
– investigate self-expression through vertical function migration– investigate self-expression through self-optimisation
ensuring correctness through self verification
9
– ensuring correctness through self-verification– ensuring reliability through thread-level fault tolerance
Networking and MiddlewareNetworking and Middlewarenetwork: threat level, congestion, error rate, locality, …
L inode: battery status, available hardware, user demands,environment, …
Learning
Self-awareness
prediction routing
application
Self-expression
Feedback monitoringsecurity
transport
Strategyprotocol graph adaptation,hw/sw thread assignment
linktransport
• research topicsp– develop autonomous networking architecture based on self-aware node technology– develop resource-aware middleware enabling horizontal function migration– investigate in-network self-calibration techniques
10
investigate in network self calibration techniques– ensure reliability through network-level fault tolerance
Engineering Proprioception in Computing SystemsEngineering Proprioception in Computing SystemsEngineering Proprioception in Computing SystemsEngineering Proprioception in Computing Systems(EPiCS)(EPiCS)
—— Project Overview Project Overview ——
M Pl tMarco [email protected]
AWARE MeetingAWARE MeetingAmsterdamDecember 14-15, 2010
OutlineOutline•• Proprioceptive Computing Systems: PCSProprioceptive Computing Systems: PCS• EPiCS Consortium• EPiCS Consortium• Applications• Concepts and Foundations• Hardware/Software PlatformHardware/Software Platform• Networking and Middleware
RECOGNITION: Relevance andRECOGNITION: Relevance and Cognition for Self‐Awareness in
a Content‐Centric Internet
Stuart M. Allen, Franco Bagnoli, Gualtiero Colombo, gMarco Conti, Jon Crowcroft, Chris Jones, Pietro Liò,
Refik Molva, Melek Onen, Andrea Passarella, k k h k k kIoannis Stavrakakis, Roger M. Whitaker, Eiko Yoneki
RECOGNITION overviewDecember 2010
1
Motivation: Technological TrendsMotivation: Technological Trends• Participatory generation of contentp y g
– Prosumers, diversity, expanding edges– Long tail, swamping, scale!
• Content in the environment– Linkage of the physical and virtual worlds– Embedding content and knowledge
• Acquiring knowledge through social q g g gmechanisms– Blogging, social networking, recommendation, RSS feeds…
• How content reaches users will ti t h
RECOGNITION overviewDecember 2010
continue to change…2
Self‐awareness to support technological trends
• Our Intention: Paradigm to support ICT f tiICT functions – Enabling content centricity
• Better fitting of users to content and vice• Better fitting of users to content and vice versa
– Synchronize content with human activity and needs
• Place, time, situation, relevance, context, social searchsocial search
– Autonomic management• Of content, its acquisition and resource
lRECOGNITION overview
December 2010
utilization
3
Human Awareness BehavioursHuman Awareness Behaviours
A h C & l i k• Approach: Capture & exploit key behaviours of the most intelligent living speciesliving species– Human capability is phenomenal in navigating complex & diverse stimulinavigating complex & diverse stimuli
– Filter & suppress information in “noisy” situations with ambient stimuli
– Extract knowledge in presence of uncertaintyE i id l j d t f– Exercise rapid value judgment for prioritisation
– Engage a social context and multi‐scale
RECOGNITION overviewDecember 2010
Engage a social context and multi scale learning
4
Human Awareness BehavioursHuman Awareness Behaviours
Cognitive psychological basisFor awareness and understanding
Defining key principles for exploitation by technology componentstechnology components
Embedding these principles for self‐awareness in autonomic content acquisition in pervasive environments
Potential change in behaviour due to self–awareness in ICT
RECOGNITION overviewDecember 2010
5
Overview of StructureOverview of Structure
UNIFI LEADCU LEAD
CNR LEAD
NKUA LEAD UCAM LEADRECOGNITION overview
December 20106
NKUA LEAD UCAM LEAD
Providing Autonomic Content Management
Th h R iti “N d ” t t b lf• Through Recognition “Nodes”, content becomes as self‐aware as devices
• Allow individuals to gain content that they didn’t know g ythey wanted…
• Geo‐Informatics: space, place, time…C t t l t & t i l b d it ti d l ti– Content placement & retrieval based on situation and location
• Storage and forwarding decisions based on relevance from:– Social contextSocial context– Location & environment
• Trust & security management– Uncertainty & belief
RECOGNITION overviewDecember 2010
7
Interdisciplinary DimensionsInterdisciplinary Dimensions
– Complex systems
– Artificial intelligence
– Geo‐informatics
– Cognitive psychology– Cognitive psychology
– Information retrieval
– Communication systems
– Security, trust
RECOGNITION overviewDecember 2010
8
Key QuestionsKey Questions…
• Psychology– What key concepts should be develop/include?y p p/
– Can these be used in different parts of the project?
• Scenarios• Scenarios– What contemporary areas of “social computing” are key to prioritise?key to prioritise?
– What would have the biggest impact?
– Are there demo’s that could be developed?
• Other questions…….
RECOGNITION overviewDecember 2010
9
Proposal: “Psychology” areasProposal: Psychology areas
b b l l d l• Recognition, Probabilistic Mental Models, Heuristics
H h t i ti f t– Human characteristics for agents– Decision making under bounded rationality
• Social Learning• Social Learning– Observing, retaining, learning, replicating (mimicking)
• Spatial Cognition• Spatial Cognition– Space, place, context
• Belief Desire and Intention models• Belief, Desire and Intention models– Pulling from different areas of psychology but not fully grounded
RECOGNITION overviewDecember 2010
grounded
10
1 Relevance Theory1 ‐ Relevance Theory• Sperber and Wilsonp
– Non‐coding model of communication– Inferential model taking into account context via “utterances”
– provide "cognitive effects" worthy of the processing effort required to find theprocessing effort required to find the meaning
• The speaker purposefully gives a clue to the hearer
• The hearer infers the intention from the clue and the context‐mediated information. The hearer must interpret the clue, taking into account the context, and surmise what the speaker intended to communicate.
RECOGNITION overviewDecember 2010
11
2 Judgment & Decision Making2‐ Judgment & Decision Making
• Work of Daniel Goldstein et al
– Heuristics that make us smart…• “Take the best” heuristic• Recognition heuristic
Bounded rationality– Bounded rationality• Limited direct knowledge/partial info• Fast inference has to be made• Fast inference has to be made…
RECOGNITION overviewDecember 2010
12
2 Judgment & Decision Making2‐ Judgment & Decision Making • Take the best heuristicTake the best heuristic
– judgment based on multiple criteria• the criteria are tried one at a time• the criteria are tried one at a time according to their “cue validity”
• high cue validity for a given feature g y gmeans that the feature or attribute is more diagnostic of the class membership than a feature with low cue validitythan a feature with low cue validity
– a decision is made based on the first discriminating criteriondiscriminating criterion
• the heuristic did well at making accurate inferences in real world environments
RECOGNITION overviewDecember 2010
13
2 Judgment & Decision Making2‐ Judgment & Decision Making • Recognition heuristicRecognition heuristic
– If one of two objects is recognized and the other is not then infer that thethe other is not, then infer that the recognized object has the higher value with respect to the criterion.p
– Sensitive to the criterion• Methodology for “cue validity”Methodology for cue validity
– Less‐is‐more effect • Limited information does not impedeLimited information does not impede performance (to the contrary!)
RECOGNITION overviewDecember 2010
14
3 Spatial Cognition3‐ Spatial Cognition• Human understanding and meaning forHuman understanding and meaning for ill‐defined but commonly used spatial terms
• South east…• South Wales• Central london
• Use of these in geo‐spatial content g pso that it can become self‐aware
RECOGNITION overviewDecember 2010
15
Key QuestionsKey Questions…
• Psychology– What key concepts should be develop/include?y p p/
– Can these be used in different parts of the project?
• Scenarios• Scenarios– What contemporary areas of “social computing” are key to prioritise?key to prioritise?
– What would have the biggest impact?
– Are there demo’s that could be developed?
• Other questions…….
RECOGNITION overviewDecember 2010
16
Candidate ScenariosCandidate Scenarios
• Information Retrieval & content provision– Human awareness when using search engine interfaces – e.g., automatic cue detection & HCI
• Self‐aware Multimedia and “Active” Data– MP3, other types of content– Self‐aware meta‐data for spatial problemsSelf‐aware meta‐data for spatial problems
• Social Computingd i d i fil i i– Crowd sourcing, recommendation, filtering, micro‐
blogging, tagging
RECOGNITION overviewDecember 2010
17
RECOGNITION: Relevance andRECOGNITION: Relevance and Cognition for Self‐Awareness in
a Content‐Centric Internet
Stuart M. Allen, Franco Bagnoli, Gualtiero Colombo, gMarco Conti, Jon Crowcroft, Chris Jones, Pietro Liò,
Refik Molva, Melek Onen, Andrea Passarella, k k h k k kIoannis Stavrakakis, Roger M. Whitaker, Eiko Yoneki
RECOGNITION overviewDecember 2010
18
SAPERE: Self‐aware Pervasive Service Ecosystems
Franco Zambonelli
Università di Modena e Reggio Emilia
STREP – 3 years
www sapere‐project euwww.sapere‐project.eu
1AWARE Meeting ‐ December 14, 2010
The ConsortiumThe Consortium
• Università di Modena e Reggio Emilia– Franco Zambonelli & Marco MameiFranco Zambonelli & Marco Mamei
• Birkbeck College University of LondonGiovanna Di Marzo Serugendo– Giovanna Di Marzo Serugendo
• Alma Mater Studiorum – Università di Bologna– Mirko Viroli & Andrea Omicini
• St. Andrews University– Simon Dobson
• Johannes Kepler Universitaet Linzp– Alois Ferscha
2AWARE Meeting ‐ December 14, 2010
The ScenarioThe Scenario
• Pervasive computing– Sensor rich and always connected smart phones – Sensor networks and information tags– Localization and activity recognitionInternet of things and the real time Web– Internet of things and the real‐time Web
• Innovative pervasive services arising– Situation‐aware adaptationSituation aware adaptation – Interactive reality – Pervasive collective intelligence and pervasive participation
• Open co‐production scenario, very dynamic, diverse needs and diverse services, continuously evolving
3AWARE Meeting ‐ December 14, 2010
The Overall ObjectiveThe Overall Objective
• Develop and demonstrate a highly‐innovative theoretical and practical framework for pervasive service ecosystemsservice ecosystems– Adaptivity and self‐management as inherent properties of the ecosystemy
– Systemic self‐awareness as an observable property of the overall systemLong lasting (eternal) adaptivity– Long‐lasting (eternal) adaptivity
– Bio‐chemical inspiration• Foundational re‐thinking ofFoundational re thinking of
– Service architectures and associated middleware– Self‐* algorithms and contextual knowledge management
4AWARE Meeting ‐ December 14, 2010
The Architectural ApproachThe Architectural Approach
• Open production model• Smooth data/services
distinction– LSA live semantic
annotations• InteractionsInteractions
– Sorts of bio‐chemical reactions among components
– In a spatial substrateIn a spatial substrate• Eco‐laws
– Rule all interactionsDiscovery + orchestration– Discovery + orchestration seamlessly merged
• Built over a pervasive network world
AWARE Meeting ‐ December 14, 2010 5
world
Specific ObjectivesSpecific Objectives
• Both of a scientific and technological nature
• Around which the various WPs are organized
AWARE Meeting ‐ December 14, 2010 6
Model Structures and KnowledgeModel, Structures, and Knowledge
• M d l & M th d l• Model & Methodology– Innovative chemical‐inspired semantic model for interactions among
components and their dynamic composition/aggregationS ti (LSA) d i ti d ti tt thi– Semantic (LSA) description and semantic pattern‐mathing
– Uniform traitment of data and services– Methodological guidelines associated
• Structures & Space– Model distributed self‐* algorithms via the chemical LSA framework– Innovative flexible means for aggregation and compositiongg g p– Define decentralized means to control the behaviour of the ecosystem
• Knowledge & Time– Distributed knowledge management algorithms via the LSA framework– Distributed knowledge management algorithms via the LSA framework– Define new means to perform distributed recognition of current situations– As well as to enable recognition of future situations
AWARE Meeting ‐ December 14, 2010 7
Key Challengesy gfor model, structure and knowledge
C h i ll i i d t ti d l d th• Can our chemically‐inspired computation model and the eco‐laws?– Be flexible and general‐purpose enough?g p p g– Effectively deal with the complexity and diversity of modern
pervasive scenarios?– Be effectively implementable?Be effectively implementable?
• And, for structure and knowledge– Can it accommodate all needed distributed aggregation and
lf iti l thiself‐composition algorthims– Can it express all needed forms of knowledge management?
• Or should we rather go for application‐specific (or location‐Or should we rather go for application specific (or locationspecific) eco‐laws?
AWARE Meeting ‐ December 14, 2010 8
Infrastructure and ApplicationsInfrastructure and Applications
• I f t t• Infrastructure– A very lightweight infrastructure– Ruling all interactions (from discovery to data exchange and
h i ti ) b b ddi th t f lsynchronization) by embedding the concept of eco‐laws– To most extent, acting as a recommendation and planning engine– Possibly inspired by tuple space coordination models– Yet made it more “fluid” and suitable for a pervasive computing
continuum substrate not a network but a continuum of tuple spaces
• Applications– The “Ecosystem of Display” as a general and impactful testbed– To put at work and demonstrate the SAPERE findingsp g– Active and dynamic information sharing in urban scenarios– Active participation of citizens to the working of the urban infrastructure
AWARE Meeting ‐ December 14, 2010 9
Key Tangible Resultsy g(hopefully)
• A novel model and methodology to support the development of complex service systems in open and dynamic pervasive scenariosscenarios
• A uniform set of:– Self‐* algorithms for service/data composition and aggregation (in
th f f lib i )the form of libraries)– Algorithms and tools for distributed management of contextual‐
knowledge, to enforce present‐ and future‐awareness in the ecosystem
• A novel middleware for pervasive computing scenarios (Open Source))– Integrating the stated algorithms in the form of libraries
• A set of released innovative application showcased on the E t f Di l t tb dEcosystem of Displays testbed
AWARE Meeting ‐ December 14, 2010 10
KEYWORDS
Awareness MeetingAwareness Meeting
AdaptationAdaptation
• Adaption on multiple timescales
• Organised adaptationOrganised adaptation
• Adaptation to hostile situations
• Adaptation to changing environments
• Adaptation for robustnessAdaptation for robustness
Evolution/EmergenceEvolution/Emergence
Evolution
• Evolution of new collective
Emergence
• Emergent Systemsbehaviours
• Open‐ended evolution• Emergent behaviours
Self *Self‐
Self‐properties
• Self‐expression
Self‐awareness
• of state
• Self‐optimisation
• Self‐organising networks
• about environment
• of contextg g
• Self‐organisation • Collective self‐awareness
• Situation awareness
Awareness of meAwareness of me
• How do others see me ?
• Look‐* self‐awarenessLook self awareness
• Is the environment aware of me ?
Learning/BehaviourLearning/Behaviour
• Learning
• CognitionCognition
• Filtering
• Characteristics of behaviours
• Opportunistic behaviourOpportunistic behaviour
• Knowledge
• Knowledge‐intensive systems
Distribution and collectivesDistribution and collectives
Distributed
• Decentralised systems
Collectives
• Collective intelligence
• Distributed artificial intelligence
• Global behaviour – local decisions
• Robust distributed systems
• Distributed Control
• Coordination technologies
• Collaborative decision making
System PropertiesSystem Properties
Robust/Resilient
• Fault tolerance
Others!
• Relevant
• Robustness to sub‐ideal operation
• Out of control
• Homeostasis• Resilience • Efficient
• Autonomous
Socially InspiredSocially Inspired
• Social media
• Social cognitionSocial cognition
• Human cognition
• Human in the loop
• Augmented societyAugmented society
• Social networking
• Socio‐technical combinatorics
Services/SystemsServices/Systems
• Adaptive middleware
• Architecture support for adaptivityArchitecture support for adaptivity
• Self‐joining services
• Common services (middleware)
• Service oriented architectureService oriented architecture
• Autonomic service components
Information and ModellingInformation and Modelling
Information/Recognition
• Introspection about norms
Models
• Meta‐modelling of run‐time and conventions
• Utility of information
behaviour
• Organisational models
• Intention recognition
• Event recognition
• Modelling the environment
• modelling inner state
Techniques/SystemsTechniques/Systems
Techniques• Bio‐inspired computing
Systems
• Multi‐agent systems• Stream computing• Software‐engineering
• Ensembles
• Self‐governing ensembles• Pervasive computing• Social computing
bili
g g
• Cloud computing
• Sensor networks• E‐mobility• Languages
• Eco system
• Robot swarms
• Measurement • Autonomous systems
QuestionsQuestions
• How does self‐awareness relate to self‐* ?
• How are self‐aware systems designed ?How are self aware systems designed ?
• What are meaningful applications of self‐?awareness ?
• How do we program such systems ?p g y
• How do we enable innovations ?