Artificial Recognition System (ARS) Project General-purpose model of - - PowerPoint PPT Presentation

artificial recognition system ars project
SMART_READER_LITE
LIVE PREVIEW

Artificial Recognition System (ARS) Project General-purpose model of - - PowerPoint PPT Presentation

www.ict.tuwien.ac.at A rtificial R ecognition S ystem Development and Evaluation Samer Schaat , Alexander Wendt, Matthias Jakubec, Friedrich Gelbard, Lukas Herret, and Dietmar Dietrich Institute of Computer Technology /15 www.ict.tuwien.ac.at


slide-1
SLIDE 1

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Artificial Recognition System

Development and Evaluation

Samer Schaat, Alexander Wendt, Matthias Jakubec,

Friedrich Gelbard, Lukas Herret, and Dietmar Dietrich

slide-2
SLIDE 2

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Artificial Recognition System (ARS) Project

2 Samer Schaat

  • General-purpose model of human information processing for

the usage in various artificial systems

  • Humanoid agents in a virtual world

Body Psyche

Decision unit (ARS model)

Human-Robot Interaction (Kismet) Evacuation Simulation (ESCAPES)

slide-3
SLIDE 3

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Key Features of the ARS Approach

3 Samer Schaat

  • Functional model

Generative approach: describing functions not behavior  generic, flexible

  • Layered description model

Appropriate means of description for different aspects (neurons, neurosymbolics, psyche)

  • Holistic and unitary model

Consistent and coherent integration of basic aspects (motivation, emotion, planning…)

  • Top-down approach

Concretize abstract functions incrementally, starting with psychic layer

  • Bionic and interdisciplinary approach

Translate knowledge into technical models

slide-4
SLIDE 4

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Basic question: How to develop and evaluate such a model?

4 Samer Schaat

slide-5
SLIDE 5

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Challenges

5 Samer Schaat

  • Restricted accessibility of mind’s functioning
  • Interdisciplinary understanding and knowledge translation
  • Complexity in description and explanation
slide-6
SLIDE 6

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Restricted Accessibility of the Mind

6 Samer Schaat

  • Various ways to get information about the mind’s functioning
  • Relevant knowledge for our objective? Right level (psyche)?
  • Cannot be used directly
  • Interpretation and knowledge

translation required  Experts needed

http://homepages.uni-tuebingen.de/karnath/Research.html http://www.edgehill.ac.uk/psychology/research-participation/ http://de.wikipedia.org/wiki/Elektroenzephalografie

slide-7
SLIDE 7

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Interdisciplinary Understanding

7 Samer Schaat http://variationsphase.de/vp/2012/10/misunderstanding/

  • Regular, intensive

collaboration

  • Different concepts,

vocabulary….?

slide-8
SLIDE 8

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Complexity and Explanation

8 Samer Schaat http://medtech-news.com/?p=38

  • Right level, relevant knowledge?
  • Not only on neuronal level, also on psychic level
  • Interplay of various factors determine behavior
slide-9
SLIDE 9

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Case-driven Agent-based Simulation

9 Samer Schaat

Combination of

  • Casuistics for interdisciplinary collaboration
  • UC-based requirement analysis for deterministic structuring
  • Agent-based simulation as a evaluation framework
slide-10
SLIDE 10

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Step 1: Describe phenomena and assumptions

10 Samer Schaat

  • Platform and tool for interdisciplinary collaboration
  • Exemplify and discuss research question with a concrete exemplary case

e.g. How two hungry agents behave in front of a food source (eat, share…)

  • Enables stating (and testing) concrete assumptions

(e.g. the role of emotions, drives, and norms)

  • Avoids drifting into abstract discussions
  • Embodies and integrate theories from different disciplines to explain

behavior

State of the art, experts‘ interpretation of real world conditions

  • But: indeterministic, gaps in assumptions, inconsistent  no direct usage
slide-11
SLIDE 11

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Step 2: Analysis and Structuring

11 Samer Schaat

  • Clarify the exemplary case
  • Explication of assumptions
  • Consistent description
  • Structure to deterministic

description

  • Causal function description
  • Data determinants of behavior

(Memories, personality, environment, internal state)

  • Simulation-case (SC) enables
  • Requirements analysis
  • Computational model
  • Test plan for evaluation
slide-12
SLIDE 12

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Step 3: Data and Functional Model

12 Samer Schaat

  • Previous steps enable
  • Requirements statement
  • Algorithmic description of functions
  • Modelling of knowledge representation
  • Specify function modules, interfaces, data

Adaption or extension?

  • Implemented in MASON (Java) and Protégé (Ontology)
slide-13
SLIDE 13

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Step 4: Evaluation

13 Samer Schaat

  • Simulation-case as test-template  parameterize simulation according to

scenarios

  • Does the functions generate and data

determine behavior as expected?

  • How is the behavior generated?
  • Test our hypotheses’ predictability
  • Are the assumptions of exemplary case valid?
  • Does the interplay of specified factors (e.g. emotions, drives, norms)

generate the expected behavior?

  • Does the specified data determine behavior (change)?
  • Unexpected behavior or state  analysis on different levels  feedback

cycles

slide-14
SLIDE 14

www.ict.tuwien.ac.at

Institute of Computer Technology /15 14 Samer Schaat

Conclusion

  • Feedback cycles
  • Possibility a, b: mistake in model translation
  • Possibility c: inconsistent in or between theories
  • Bridge disciplines, test knowledge translation
  • Concretize testable assumptions

from other disciplines

  • Structure interdisciplinary knowledge to a

causal model and test plan

  • SC scenarios  model calibration
  • Stable model?  sensitivity analysis!
  • Premises for model application in

specific domains  Outlook

slide-15
SLIDE 15

www.ict.tuwien.ac.at

Institute of Computer Technology /15

Thank you!

15 Samer Schaat