Explanation in GILA 2 Stanford -> RPI McGuinness, Ding January 15, 2008.

Slides:



Advertisements
Similar presentations
Geoinformatics 2008 Fox Semantic Provenance 1 Semantic Provenance for Image Data Processing Peter Fox (HAO/ESSL/NCAR) Deborah McGuinness (RPI) Jose Garcia,
Advertisements

L3S Research Center University of Hanover Germany
Solving inconsistent ontologies with heuristics Joey Lam, Derek Sleeman, Wamberto Vasconcelos 23 Jan 2006.
GMD German National Research Center for Information Technology Darmstadt University of Technology Perspectives and Priorities for Digital Libraries Research.
Translation-Based Compositional Reasoning for Software Systems Fei Xie and James C. Browne Robert P. Kurshan Cadence Design Systems.
Search in Source Code Based on Identifying Popular Fragments Eduard Kuric and Mária Bieliková Faculty of Informatics and Information.
Model Checker In-The-Loop Flavio Lerda, Edmund M. Clarke Computer Science Department Jim Kapinski, Bruce H. Krogh Electrical & Computer Engineering MURI.
Programming Types of Testing.
K S L W i n e A g e n t : Testbed Application for Semantic Web Technologies Deborah McGuinness Eric Hsu Jessica Jenkins Rob McCool Sheila McIlraith Paulo.
Selecting Preservation Strategies for Web Archives Stephan Strodl, Andreas Rauber Department of Software.
Expert System Human expert level performance Limited application area Large component of task specific knowledge Knowledge based system Task specific knowledge.
Chapter 6: Design of Expert Systems
McGuinness – Microsoft eScience – December 8, Semantically-Enabled Science Informatics: With Supporting Knowledge Provenance and Evolution Infrastructure.
Knowledge Acquisitioning. Definition The transfer and transformation of potential problem solving expertise from some knowledge source to a program.
Understanding Metamodels. Outline Understanding metamodels Applying reference models Fundamental metamodel for describing software components Content.
1 Chapter 9 Rules and Expert Systems. 2 Chapter 9 Contents (1) l Rules for Knowledge Representation l Rule Based Production Systems l Forward Chaining.
Explaining Preference Learning Alyssa Glass CS229 Final Project Computer Science Department, Stanford University  Augment PLIANT to gather additional.
Automatically Extracting and Verifying Design Patterns in Java Code James Norris Ruchika Agrawal Computer Science Department Stanford University {jcn,
EXPERT SYSTEMS Part I.
Programming Fundamentals (750113) Ch1. Problem Solving
1 The Problem o Fluid software cannot be trusted to behave as advertised unknown origin (must be assumed to be malicious) known origin (can be erroneous.
PML Model for GILA James Michaelis 11 / 13 / 2008.
Explanation and Trust for Adaptive Systems Alyssa Glass (Stanford / SRI / Willow Garage) In collaboration with Deborah McGuinness (Stanford/RPI), Michael.
LÊ QU Ố C HUY ID: QLU OUTLINE  What is data mining ?  Major issues in data mining 2.
Semantic Interoperability Jérôme Euzenat INRIA & LIG France Natasha Noy Stanford University USA.
Špindlerův Mlýn, Czech Republic, SOFSEM Semantically-aided Data-aware Service Workflow Composition Ondrej Habala, Marek Paralič,
Web Explanations for Semantic Heterogeneity Discovery Pavel Shvaiko 2 nd European Semantic Web Conference (ESWC), 1 June 2005, Crete, Greece work in collaboration.
Katanosh Morovat.   This concept is a formal approach for identifying the rules that encapsulate the structure, constraint, and control of the operation.
© Janice Regan, CMPT 128, Jan CMPT 128 Introduction to Computing Science for Engineering Students Creating a program.
March 26, 2007 McGuinness et al Explaining Task Processing in Cognitive Assistants that Learn Deborah McGuinness 1, Alyssa Glass 1,2, Michael Wolverton.
Investigations into Trust for Collaborative Information Repositories: A Wikipedia Case Study Deborah L. McGuinnessDeborah L. McGuinness, Co-Director Knowledge.
Mihir Daptardar Software Engineering 577b Center for Systems and Software Engineering (CSSE) Viterbi School of Engineering 1.
Explanation: The Next Phase in Question Answering Deborah L. McGuinness Knowledge Systems Laboratory Stanford University
PART IV: REPRESENTING, EXPLAINING, AND PROCESSING ALIGNMENTS & PART V: CONCLUSIONS Ontology Matching Jerome Euzenat and Pavel Shvaiko.
Semantic Learning Instructor: Professor Cercone Razieh Niazi.
Carnegie Mellon School of Computer Science Copyright © 2001, Carnegie Mellon. All Rights Reserved. JAVELIN Project Briefing 1 AQUAINT Phase I Kickoff December.
Supporting Civil-Military Information Integration in Military Operations Other than War Paul Smart, Alistair Russell and Nigel Shadbolt
© DATAMAT S.p.A. – Giuseppe Avellino, Stefano Beco, Barbara Cantalupo, Andrea Cavallini A Semantic Workflow Authoring Tool for Programming Grids.
Inference Web: Portable and Sharable Proofs for Hybrid Systems Deborah L. McGuinness, Paulo Pinheiro da Silva and Bill MacCartney with Richard Fikes, Gleb.
CONCLUSION & FUTURE WORK Normally, users perform search tasks using multiple applications in concert: a search engine interface presents lists of potentially.
Introduction to Tetherless World RPI by Jie Bao Slides will be available from:
Indirect Supervision Protocols for Learning in Natural Language Processing II. Learning by Inventing Binary Labels This work is supported by DARPA funding.
1 Semantic Provenance and Integration Peter Fox and Deborah L. McGuinness Joint work with Stephan Zednick, Patrick West, Li Ding, Cynthia Chang, … Tetherless.
Personalized Interaction With Semantic Information Portals Eric Schwarzkopf DFKI
1 Foundations VI: Provenance Deborah McGuinness and Peter Fox CSCI Week 12, November 30, 2009.
123 Jiao Tao 1, Li Ding 2, Deborah L. McGuinness 3 Tetherless World Constellation Rensselaer Polytechnic Institute Troy, NY, USA 1 PhD Student 2 Postdoctoral.
Enabling Explanations: The Inference Web and PML Approach Deborah McGuinness, Paulo Pinheiro da Silva, Li Ding Knowledge Systems Laboratory Stanford University.
Tool for Ontology Paraphrasing, Querying and Visualization on the Semantic Web Project By Senthil Kumar K III MCA (SS)‏
Realities in Science Data and Information - Let's go for translucency AGU FM10 IN13B-02 Peter Fox (RPI) Tetherless World.
Dictionary based interchanges for iSURF -An Interoperability Service Utility for Collaborative Supply Chain Planning across Multiple Domains David Webber.
A Semantic Web Approach for the Third Provenance Challenge Tetherless World Rensselaer Polytechnic Institute James Michaelis, Li Ding,
Explainable Adaptive Assistants Deborah L. McGuinness, Tetherless World Constellation, RPI Alyssa Glass, Stanford University Michael Wolverton, SRI International.
Review of Parnas’ Criteria for Decomposing Systems into Modules Zheng Wang, Yuan Zhang Michigan State University 04/19/2002.
Personalized Recommendation of Related Content Based on Automatic Metadata Extraction Andreas Nauerz 1, Fedor Bakalov 2, Birgitta.
Explainable Adaptive Assistants Deborah L. McGuinness, Tetherless World Constellation, RPI Alyssa Glass, Stanford University Michael Wolverton, SRI International.
Explanation Infrastructure Supporting Transparency and Accountability Deborah L. McGuinness Co-Director and Senior Research Scientist Knowledge Systems,
Artificial Intelligence
Concept mining for programming automation. Problem ➲ A lot of trivial tasks that could be automated – Add field Patronim on Customer page. – Remove field.
© The ATHENA Consortium. CI3 - Practices of Interoperability in SMEs Proposed Solutions.
Metadata Driven Aspect Specification Ricardo Ferreira, Ricardo Raminhos Uninova, Portugal Ana Moreira Universidade Nova de Lisboa, Portugal 7th International.
Business Monitoring Framework for Process Discovery with Real-Life Logs Mari Abe Michiharu Kudo IBM Research - Tokyo.
Of 24 lecture 11: ontology – mediation, merging & aligning.
Annotating and Embedding Provenance in Science Data Repositories to Enable Next Generation Science Applications Deborah L. McGuinness.
Towards a framework for architectural design decision support
Encoding Extraction as Inferences
Architecture Components
Ontology Evolution: A Methodological Overview
Chapter 5 Designing the Architecture Shari L. Pfleeger Joanne M. Atlee
Understanding PML A Proof Markup Language
Foundations VI: Provenance
Presentation transcript:

Explanation in GILA 2 Stanford -> RPI McGuinness, Ding January 15, 2008

Motivation Improve trust in recommendations from GILA components Support evaluation – Understand why GILA makes suggestions – Identify which prior knowledge is (re-)used – Identify which learned knowledge is learned and (re-)used – Summarize usage of external interaction information steps Support internal trust and reuse – Identify which component suggested what and why – Identify/ propagate dependencies

Provided Knowledge runtime input prior input Learned Knowledge Final Output Flow of GILA Knowledge expert execution trace ontological knowledge constraint-violation pstep list as the final solution Learning & practice performance context knowledge facts embedded in the input problem problem/solution constraints conflict priority

GILA Knowledge KnowledgeCategoryProducerConsumerOWL class Execution trace ProvidedExpertMRE, ILRsgilcore:ExecutionTrace Input ACO problem ProvidedBlueforceMRE, ILRsgilaco:ProblemACO Context knowledge ProvidedContext service MRE, ILRsPhase II Ontological knowledge ProvidedExpertMRE, ILRsaco:* Constraint LearnedCLILRsgilcore:Constraint Constraint Violation LearnedSCMRE, ILRsgilaco:ViolationStatement An ACO problem state LearnedMREILRsgilaco:ProblemACO Cost of an ACO problem state LearnedDTLMRE, ILRs?gilaco:CostStatement Credit/Blame Assignment LearnedSC + MREILRsPhase II Conflicts in an ACO problem state Learned4DMREgilcore:Conflict Order of conflicts LearnedILRsMREPhase II Intersection of conflicts Learned4DMREgkst:IntersectionDetails Pseudo expert trace LearnedMRE, ILRsILRsgilcore:ExecutionTrace Final solution LearnedMRE-DMPstepManagergilcore:PstepList

Phase II Objectives Show usage of provided knowledge – Expert execution trace – Ontological knowledge Show usage of learned knowledge – Constraints – Blame (constraint violation) assignments Show (abstracted) problem-solving trace (with dependencies) – problem, learner, solution, …., final solution

Approach Ontology Representation – Reference knowledge at appropriate level of granularity – Provide interlingua for learners’ knowledge usage Extract some knowledge from KB Derive new knowledge from existing knowledge Computational Components – Run time API for sharing knowledge of explanations – API for extracting, summarizing and visualizing problem- solving and knowledge usage trace utilizing Inference Web approach

Example: Blame-assignment #vstmt192 (gilaco:ViolationStatement) Degree of importance of violation How severe the violation is Confidence #constraint 7 (gilcore:Constraint) #pstep261 (gilcore:Pstep) #problem26 (gilaco:ProblemTrySafetyConstraint) #solution28 (gilaco:SolutionTrySafetyConstraint) Safety Checker MRE-DM #solution23 (gilaco:SolutionResolveConflict) #solution26 (gilaco:SolutionResolveConflict) X-ILR DT-ILR …

Learned knowledge Provided knowledge Example: Learn and Solve Based on correspondence, santi Nov 30, 2007 Solution 5647Case 101 PSTEP 8859 PSTEP 8860 PSTEP 8861 ExecutionTrace 1878 generarted from learned from learned from learned from in confidence confidence Performance Step 129 Learning Step 126 CBL ILR Has conclusion Has antecedent Has conclusion Has antecedent match case and solve Learn new case Has learner Use rule

Directions Determine appropriate granularity for representation and propagation (initially coarse level moving to finer granularity where required) Design appropriate primitives for phase II topics – context, credit/blame, orderings, priorities, … Focus on dependencies initially (supporting explanations showing usage of prior knowledge, external interaction- gained info, use and re-use of learned knowledge, similarity knowledge, adaptation knowledge…) Design GILA-appropriate explanation templates exploiting our explanation interlingua Present knowledge provenance summaries (with follow up options)

PSTEP Manager Fujitsu Song

PSTEP Manager in Phase 1 SWebAD PSTEP Manager (PMAN) PSTEP List Translate into SWebAD command Backup PSTEP information Execution Generate execution trace Known Issues – PMAN is widely used to check whether a proposed solution is correct or not, each check takes long time. – Most failures are due to some errors which can be checked before execution and avoided (e.g. NaN maximum altitude, minimum altitude > maximum altitude, etc.) – The knowledge may not be discovered from the expert trace (since the expert does not make this type of mistakes), but can be learned from the execution result reported by SWebAD during the execution time. PSTEP Manager in Phase 1 is a simple execution engine and it hides execution related details from other modules

PSTEP Manager in Phase 2 STBMCS PSTEP List Translate into STBMCS command Backup PSTEP information Generate execution trace Early Error Detection Execution Request STBMCS Error Learner Constraint knowledge repository PSTEP List Optimization New Modules Existing Modules PSTEP Manager (PMAN) STBMCS Error learner learns new knowledge from error report of STBMCS (previously SWebAD) Share constraint knowledge with 4DCL Use constraint knowledge to detect errors of proposed solution before real execution Share with 4DCL/R Error Report

Extra

Files/WWW Toolkit Proof Markup Language (PML) CWM (NSF TAMI) JTP (DAML/NIMD) SPARK (DARPA CALO) UIMA (DTO NIMD Exp Aggregation) IW Explainer/ Abstractor IWBase IWBrowser IWSearch Trust Justification Provenance N3 KIF SPARK-L Text Analytics IWTrust provenance registration search engine based publishing Expert friendly Visualization End-user friendly visualization Trust computation Semantic Discovery Service (DAML/SNRC) OWL-S/BPEL Framework for explaining question answering tasks by abstracting, storing, exchanging, combining, annotating, filtering, segmenting, comparing, and rendering proofs and proof fragments provided by question answerers. Inference Web Infrastructure McGuinness, Ding, Pinheiro da Silva, Chang, Fikes, Glass, Zeng