ARCHITECTURES AND STANDARDS FOR IVAS AT THE SOCIAL COGNITIVE SYSTEMS GROUP H. van Welbergen, K. Bergmann, H. Buschmeier, S. Kahl, I. de Kok, A. Sadeghipour,

Slides:



Advertisements
Similar presentations
Cultural Heritage in REGional NETworks REGNET Review Meeting (REV-01-01), , Brussels.
Advertisements

CONCEPTUAL WEB-BASED FRAMEWORK IN AN INTERACTIVE VIRTUAL ENVIRONMENT FOR DISTANCE LEARNING Amal Oraifige, Graham Oakes, Anthony Felton, David Heesom, Kevin.
GMD German National Research Center for Information Technology Darmstadt University of Technology Perspectives and Priorities for Digital Libraries Research.
National Technical University of Athens Department of Electrical and Computer Engineering Image, Video and Multimedia Systems Laboratory
INTRODUCTION TO SIMULATION WITH OMNET++ José Daniel García Sánchez ARCOS Group – University Carlos III of Madrid.
ARCHITECTURES FOR ARTIFICIAL INTELLIGENCE SYSTEMS
Component Oriented Programming 1 Chapter 2 Theory of Components.
Breakout session B questions. Research directions/areas Multi-modal perception cognition and interaction Learning, adaptation and imitation Design and.
Main issues: • Why is reuse so difficult • How to realize reuse
Herwin van Welbergen, Yuyu Xu, Marcus Thiebaux, Wei-Wen Feng, Jingqiao Fu, Dennis Reidsma, Ari Shapiro.
ENTERFACE’08 Multimodal Communication with Robots and Virtual Agents.
University of Minho School of Engineering Centre ALGORITMI Uma Escola a Reinventar o Futuro – Semana da Escola de Engenharia - 24 a 27 de Outubro de 2011.
MediaHub: An Intelligent Multimedia Distributed Hub Student: Glenn Campbell Supervisors: Dr. Tom Lunney Prof. Paul Mc Kevitt School of Computing and Intelligent.
Jenkins — Modular Perception and Control Brown Computer — ROUGH DRAFT ( ) 1 Workshop Introduction: Modular Perception.
Herwin van Welbergen Dennis Reidsma Stefan Kopp.  Beyond turn taking interaction ◦ Continuous perception and behavior generation  Interpersonal coordination.
Asa MacWilliams Lehrstuhl für Angewandte Softwaretechnik Institut für Informatik Technische Universität München Dec Software.
Effective Coordination of Multiple Intelligent Agents for Command and Control The Robotics Institute Carnegie Mellon University PI: Katia Sycara
Yaochu Jin FTR/HRE-D August, From Interactive Evolutionary Algorithms to Agent-based Evolutionary Design Interactive Evolutionary Algorithm –When.
University of Leeds Department of Chemistry The New MCM Website Stephen Pascoe, Louise Whitehouse and Andrew Rickard.
11.1 Lecture 11 CASE tools IMS Systems Design and Implementation.
SPECIFYING COGNITIVE MODELS Using Patterns and Conflicts A. Macklem, F. Mili Oakland University S. Dungrani TARDEC June, 2004.
JACK Intelligent Agents and Applications Hitesh Bhambhani CSE 6362, SPRING 2003 Dr. Lawrence B. Holder.
Provisional draft 1 ICT Work Programme Challenge 2 Cognition, Interaction, Robotics NCP meeting 19 October 2006, Brussels Colette Maloney, PhD.
Improving UML Class Diagrams using Design Patterns Semantics Shahar Maoz Work in Progress.
CSD 5230 Advanced Applications in Communication Modalities 7/3/2015 AAC 1 Introduction to AAC Orientation to Course Assessment Report Writing.
Supplement 02CASE Tools1 Supplement 02 - Case Tools And Franchise Colleges By MANSHA NAWAZ.
Chapter 10: Architectural Design
Software Architecture. Agenda " Why architect? " What is architecture? " What does an architect do? " What principles guide the process of architecting?
31 st October, 2012 CSE-435 Tashwin Kaur Khurana.
A Semantic Workflow Mechanism to Realise Experimental Goals and Constraints Edoardo Pignotti, Peter Edwards, Alun Preece, Nick Gotts and Gary Polhill School.
Mobile Multimodal Applications. Dr. Roman Englert, Gregor Glass March 23 rd, 2006.
ECSE 604: Early Literacy and Augmentative Communication, Summer 2015 Day 4: June 17 th, 2015.
The Pursuit for Efficient S/C Design The Stanford Small Sat Challenge: –Learn system engineering processes –Design, build, test, and fly a CubeSat project.
The Design Discipline.
UML - Development Process 1 Software Development Process Using UML (2)
Requirements Analysis
GUI: Specifying Complete User Interaction Soft computing Laboratory Yonsei University October 25, 2004.
Designing Software for Elderly Suffering from Cognitive Impairments M. PINO, M. BOULAY, AS. RIGAUD Research team EA 4468 Paris Descartes University, Broca.
AsapRealizer 2.0: The Next Steps in Fluent Behavior Realization for ECAs Herwin van Welbergen, Ramin Yaghoubzadeh, Stefan Kopp Social Cognitive Systems.
Affective Interfaces Present and Future Challenges Introductory statement by Antonio Camurri (Univ of Genoa) Marc Leman (Univ of Gent) MEGA IST Multisensory.
Lecture 9: Chapter 9 Architectural Design
Object-Oriented Software Engineering Practical Software Development using UML and Java Chapter 1: Software and Software Engineering.
OBJECT ORIENTED SYSTEM ANALYSIS AND DESIGN. COURSE OUTLINE The world of the Information Systems Analyst Approaches to System Development The Analyst as.
Oct 14, 2001OOPSLA’01- DSVL1 Experiences with Visual Programming Languages for End-Users and Specific Domains Philip T. Cox Trevor J. Smedley Dalhousie.
Spoken Dialog Systems and Voice XML Lecturer: Prof. Esther Levin.
GRASP: Designing Objects with Responsibilities
1 USC INFORMATION SCIENCES INSTITUTE CAT: Composition Analysis Tool Interactive Composition of Computational Pathways Yolanda Gil Jihie Kim Varun Ratnakar.
University of Windsor School of Computer Science Topics in Artificial Intelligence Fall 2008 Sept 11, 2008.
DARPA ITO/MARS Project Update Vanderbilt University A Software Architecture and Tools for Autonomous Robots that Learn on Mission K. Kawamura, M. Wilkes,
Human Factors In Visualization Research Melanie Tory and Torsten Moller Ajith Radhakrishnan Nandu C Nair.
Österreichisches Forschnungsinstitut für Artificial Intelligence Representational Lego for ECAs Brigitte Krenn.
Software Architecture Evaluation Methodologies Presented By: Anthony Register.
Design Reuse Earlier we have covered the re-usable Architectural Styles as design patterns for High-Level Design. At mid-level and low-level, design patterns.
Dialog Design I Basic Concepts of Dialog Design. Dialog Outline Evaluate User Problem Representations, Operations, Memory Aids Generate Dialog Diagram.
Multimodal Plan Representation for Adaptable BML Scheduling Dennis Reidsma, Herwin van Welbergen, Job Zwiers.
Number: TR/06/B/F/PP/ WASTE-TRAIN VOCATIONAL TRAINING, EDUCATION, CONVEYING INFORMATION ON UP-TO-DATE WASTE MANAGEMENT PRACTICES TO DECISION MAKERS/STAFF.
BID Initiative Learning Network (BLN) Update for BLN Discussion Meeting Arusha, Tanzania, 8 th December, 2015.
OASIS ebXML Registry Standard Open Forum 2003 on Metadata Registries 10:30 – 11:15 January 20, 2003 Kathryn Breininger The Boeing Company Chair, OASIS.
Understanding Naturally Conveyed Explanations of Device Behavior Michael Oltmans and Randall Davis MIT Artificial Intelligence Lab.
Seminar on SDS Final class 5/4/2015. Topics discussed ASR, NLU in dialogue Dialogue management NLG, information presentation Evaluation Error recovery.
WP6 Emotion in Interaction Embodied Conversational Agents WP6 core task: describe an interactive ECA system with capabilities beyond those of present day.
1 Usability Analysis n Why Analyze n Types of Usability Analysis n Human Subjects Research n Project 3: Heuristic Evaluation.
Software Design. Introduction Designing engineering encompasses the set of principles concepts and practices that lead to the development of a high quality.
© 2005 Prentice Hall, Decision Support Systems and Intelligent Systems, 7th Edition, Turban, Aronson, and Liang 6-1 Chapter 6 Decision Support System Development.
NCP meeting Jan 27-28, 2003, Brussels Colette Maloney Interfaces, Knowledge and Content technologies, Applications & Information Market DG INFSO Multimodal.
Augmentative and Alternative Communication Assessment and Intervention
I love portfolio! Nelly Zafeiriadou MA, EdD ELT School Advisor
Program comprehension during Software maintenance and evolution Armeliese von Mayrhauser , A. Marie Vans Colorado State University Summary By- Fardina.
Chapter 5 Designing the Architecture Shari L. Pfleeger Joanne M. Atlee
Multimodal Human-Computer Interaction New Interaction Techniques 22. 1
Presentation transcript:

ARCHITECTURES AND STANDARDS FOR IVAS AT THE SOCIAL COGNITIVE SYSTEMS GROUP H. van Welbergen, K. Bergmann, H. Buschmeier, S. Kahl, I. de Kok, A. Sadeghipour, R. Yaghoubzadeh, S. Kopp

MAIN RESEARCH THEMES  How can cognitive systems be designed as intelligent, socially adept interaction partners that allow a fluent and coordinated interaction with humans  Developing methods to model the behavioral, preceptual motor and cognitive mechanisms of embodied human-like communication and cooperation  Application and evaluation in IVAs  Scenarios:  Embedding IVAs in mouse-keyboard interaction  Virtual coaches  Virtual assistents for the elderly and cognitively impaired  Cognitive models that investigate semantic speech/gesture coordination  Computational models of dialog coordination based on linguistic feedback

CURRENT ARCHITECTURE AND STANDARDS  Our components:  A multimodal memory component realized as a spreading activation model of semantic coordination for speech-gesture production (Bergmann et al., 2013)  Incremental NLG based on SPUD (Buschmeier et al., 2012)  A Behavior Planner for iconic gestures (Bergmann and Kopp, 2009)  A BML 1.0 Realizer for incremental and adaptive behavior (van Welbergen et al. 2014)  An IS-based incremental dialog manager, capable of handling uncertain input (yet unpublished)  Several external components  Often multiple alternatives available, allowing trade-offs between recognition/synthesis quality, reactivity and control

CURRENT ARCHITECTURE AND STANDARDS  Combining components  We follow the SAIBA reference architecture  BML 1.0 for realization, extensions for incrementality and adaptivity  No standardized communication (via FML) between the Intent and Behavior Planner  Components  Run on various programming languages, OSs, computers  Allow incremental processing  IPAACA  Implements the Incremental Unit architecture (Schlangen and Skantze, 2011)  Embeds it in a message oriented middleware (RSB; Wienke and Wrede 2011)

FUTURE ARCHITECTURES AND STANDARDS: LONG TERM  Base our architecture on universal (less problem specific) and cognitively motivated principles  Ongoing work:  Fully incremental production an recognition and production, e.g. in attentive speaking (Buschmeier et al 2012)  Representation and decision making under uncertainty  Using uncertainty as valuable information rather than as a mere nuisance  Priming and imitation learning of gesture (Sadeghipour and Kopp, 2011)

FUTURE ARCHITECTURES AND STANDARDS: SHORT TERM  SAIBA provides us:  A common terminology for large scale IVA components  A standardized interface for behavior realization  We would like to enhance it to a full reference architecture for IVAs  Our requirements  Handle both input processing and output generation  Coordination between these at multiple levels  Incremental input and output processing  Representing memory  Representing uncertainty

FUTURE ARCHITECTURES AND STANDARDS Asap(Kopp et al., 2014)

SUGGESTIONS FOR DISCUSSION: A NEW REFERENCE ARCHITECTURE FOR IVAS  Our requirements  Handle both input processing and output generation  Coordination between these at multiple levels  Incremental input and output processing  Representing memory  Representing uncertainty  Requirements of others  Representing and interacting with ‘the world’  Generalization to multi-agent/multi-human

SUGGESTIONS FOR DISCUSSION: A NEW REFERENCE ARCHITECTURE FOR IVAS  Desired outcome:  Prioritized list of requirements, plan to incrementally embed them  Definition of shared terminology  Architecture sketch, identification of shared interfaces  Agenda for shared interfaces between modules  Workshops/plans of progress for BML, FML, PML, …

SUGGESTIONS FOR DISCUSSION: SHARING AND COMBINING SMALLER COMPONENTS  Many interesting ‘small’ IVA components are presented every year  In isolated projects/experiments  How can these be embedded in the larger effort of developing a full IVA?  We are especially interested in doing this for the Behavior Planner  The challenge is not only in implementing individual parts that are potentially useful but also in combining the components

SUGGESTIONS FOR DISCUSSION: SHARING AND COMBINING SMALLER COMPONENTS  Desired outcome:  (Start of) design pattern catalogue for the design of (parts of) a behavior planner  E.g. MAS, blackboard, MOM, IU-architecture, subsumption architecture  Ideas to share smaller components  List of good candidate components for sharing  Inspired by best practices from related fields (e.g. ROS: Quigley et al., 2009)  Can we develop standardized components using challenges for each of them?

SUGGESTIONS FOR DISCUSSION: CHALLENGES  Goal:  Foster the development and comparison of reusable IVA components  Different from the goals of IVA’s GALA  Provide a forum to present the state-of-the art in technology of virtual humans and innovative applications  Aimed to stimulate (under)graduate student participation  Disseminate demos in an ever evolving gallery for study and reuse  In practice the “dissemination” was limited to a video

SUGGESTIONS FOR DISCUSSION: CHALLENGES  Desired outcome:  GALA post-mortem  Best practices of successful challenges in related fields  Blizzard TTS, Give NLG, INTERSPEECH, …  Requirements/guidelines for a challenge  Plan to embed challenge in a workshop/conference (IVA?)  Example design (sketch) of a first challenge

THANKS FOR YOUR ATTENTION  Questions?

SUGGESTIONS FOR DISCUSSION: CHALLENGES  Challenges in related fields:  Blizzard TTS, Give NLG, INTERSPEECH, …  Some interesting ideas from these:  Shared set of ‘resources’, fosters comparable system building  E.g. recorded voices to build off (Blizzard), shared environment to navigate through (Give, shared corpus (INTERSPEECH)  Given by the organizers of the challenge  Predefined set of automatically obtained evaluation criteria (GIVE, INTERSPEECH)  Baselines (INTERSPEECH)  ‘Sharing’ subjects (Blizzard,  Both the challenge and the submissions are full-fledged, peer-reviewed papers (INTERSPEECH challenges, Blizzard challenges)  Mandatory (but delayed) publishing of source code and technical documentation

SUGGESTIONS FOR DISCUSSION: SHARING AND COMBINING SMALLER COMPONENTS  Inspiration from related fields: ROS (Quigley et al., 2009)  Provides over 3000 robotics components  Provides the ‘plumbing’ of robotic systems  Connecting components:  Publish subscribe, rpc, actions, distributed parameter setting, IDL’s for messages, starting components, central time, …  Debugging connections: visualization, recording/playback  Building and distributing components:  Tools for building and distributing ROS-packages  We have reinvented the wheel on several of these…  Some adhering to conventions required