Case 1:14-cv-00598-LPS Document 1-3 Filed 05/12/14 Page 54 of 110 PageID #: 72
`Case 1:14—cv—OO598—LPS Document 1-3 Filed 05/12/14 Page 54 of 110 Page|D #: 72
`
`Exh.1002
`
`E—Watch, Inc
`Exhibit 2013
`Petitioner — Iron Dome LLC
`Patent Owner — E—Watch Inc
`
`|PR2014—00439
`
`

`
`
`
`1-4C'Ow........«..a‘-'*
`
`
`
`Case 1:14-cv-00598-LPS Document 1-3 Filed 05/12/14 Page 55 of 110 PageID #: 73
`Case 1:14—cv—OO598—LPS Document 1-3 Filed 05/12/14 Page 55 of 110 Page|D #: 73
`I
`1
`
`.2
`
`Proceedings of the
`Second International Conference 0“
`Autonomous Agents
`
`,__,__....
`
`.1A,,....——-
`
`Minneapolislst. Paul, MN USA
`May 9-13, 1998
`
`Edited by Katia P. Sycara and Michael Wooldridge
`
`
`
`Iron Dome, Exh. 1002
`
`

`
`J
`
`Case 1:14-cv-00598-LPS Document 1-3 Filed 05/12/14 Page 56 of 110 PageID #: 74
`Case 1:14—cv—OO598—LPS D
`-
`ocument 1-3 % Filed 05/12/14 Page 56 of 110 Page|D #: 74
`
`The Association for Computing Machinery
`ISIS Bruadway
`New York New York 10036
`
`tputing Machinery. Inc_(ACM). Permission to
`Copyright
`I998 by the Association for Con lassroom use is granted without fee provided
`make digital or hard copies for personal or c
`that the copies are not made or distributed for profit or commercial advantage and that
`copies bear this notice and the full citation on the first page. Copyrights for components of
`this work owned by others than ACM must be honored. Abstracting with credit is permitted.
`To copy otherwise, to republish, to post on servers or to redistribute to lists requires prior
`specific permission andlor a fee. Request permission to republish from: Publications Dept.
`ACM. Inc. Fax +1 (212) 869—048l or <permissions@acn1.org> For other copying of articles
`that carry a code at the bottom of the first or last page. copying is permitted provided that the
`per-copy Fee indicated in the code is paid through the Copyright Clearance Center. 222
`Rosewood Drive. Danvers. MA 01923.
`
`ACM ISBN: ll-897‘) I -‘)83- i
`
`Additional copies may be ordered prepaid from:
`
`ACM Order Department
`PO Box lillil
`1
`Church Street Station '
`New York..,NY 10257
`
`-
`
`Phone: 1-8ti0-34236626
`(US and Canada)
`+1-212-626-0500
`(all other countries)
`Fax‘. +1-212-944-1313
`E-mail: aclnpubs({iiaent.org
`
`ACM Ortler Nutnltcr: 6t)S‘)8l
`Printed in the USA
`
`
`
`Exh. p. 2
`
`

`
`
`
`......-...._..._........-...._-
`
`
`
`
`
`Case 1:14-cv-00598-LPS Document 1-3 Filed 05/12/14 Page 57 of 110 PageID #: 75
`Case 1:14-cv-00598-LPS Document 1 3
`Filed 05/12/14 Page 57 of 110 Page|[) #; 75
`
`CONTENTS
`
`ix
`
`x
`
`Introduction
`Kaila P. Sycara. Conference Chair
`Michael Wooidridge, Program Co-Chair
`
`Conference Organizationlsponsors
`
`xi
`
`Program Committee
`
`PAPER SESSION 1: AGENTS AT THE
`INTERFACE
`Monday, May 1]
`10:30 am - 12:30 pm
`1
`Using Explicit Requirements and Metrics for
`Interface Agent User Model Correction
`Scott M. Brown. Air Force Institute of Technology
`Eugene Santos IL, University of Connecticut
`Shelia B. Banks. Air Force institute of Technology
`Mark E. Oxley. Air Force Insitute of Technology
`
`8
`
`16
`
`24
`
`Animated Autonomous Personal
`Representatives
`Timothy W. Biekmorc. FX Palo Alto Laboratory
`Linda K. Cook. ISA!
`Elizabeth F. Churchill, FX Palo Alto Laboratory
`Joseph W. Sullivan. FX Palo Alto Laboratory
`Real-Time Decision Making in Mnltimodal
`Face-to-Face Communication
`Kristinn R. Th6risson. Massachusetts Institute of
`Technology
`
`Do the Thing Right: An Architecture for
`Action-Expression
`Phoebe Sengers. Carnegie Mellon University
`
`47
`
`5 4
`
`Frontier-Based Exploration Using Multiple
`Robots
`Brian Yamauchi. Naval Research Laboratory
`
`Reconfigurable Physical Agents
`Masahiro Fajita. Sony Corporation
`Hiroaki Kitano. Sony Computer Science Laboratory
`Koji Kageyama. Sony Corporation
`
`PAPER SESSION 3: COORDINATION &
`COOPERATION
`Monday. May it
`10:30 am - 12:30 pm
`
`62
`
`Coordinating with Obligations
`Mihai Barbuceanu. University of Toronto
`Tom Gray. Mitel Corporation
`Serge Mankovski. Mite! Corporation
`
`70 Markov Tracking for Agent Coordination
`Richard Washington. Caelum Research. NASA Antes
`_ Research Center
`
`78
`
`86
`
`The CMUnited-97 Robotic Soccer Team:
`Manuela Veloso. Carnegie Mellon University
`Peter Stone. Carnegie Mellon University
`Kwun Han. Carnegie Mellon University
`
`Using Decision Tree Confidence Factors for
`Mniti-Agent Control
`Peter Stone. Carnegie Mellon University
`Manuela Veloso. Carnegie Mellon University
`
`PAPER SESSION 4: ENGINEERING AGENT
`SYSTEMS I
`Monday, May ii
`
`2:00 pm - 3:30 pm -
`
`PAPER SESSION 2: ROBOTICS
`Monday. May it
`10:30 am - i2:30 pm
`
`92
`
`32
`
`From Theory to Practice: The UTEP Robot
`in the AAAI 96 and AAAI 9'1 Robot Contests
`
`L. FIoriano,.University of Texas at El Paso
`A. Hardesty. University ofTexas at El Paso
`.
`University of Texas at El Paso
`M. Nogueira. University ofTexas at El Paso
`T. C. Son. University of Texas at El Paso
`
`39 Multi-Robot Path-Planning Based on
`Implicit Cooperation in o Robotic Swarm
`Guillaume Beslon. INSA/IF - PRISMH
`Prédérique Biennier. INSA/IF - PRlSMa
`Béat Hirsbrunncr. NUF-PA!
`
`Intelligent ond Mobile Agents
`Patterns of
`Elizabeth A. Kendall. Royal Melbourne Institute of
`Technology
`P. V. Murali Krishna. Royal Melbourne Institute of
`Technology
`Chirag V. Pathak. Royal Melbourne Institute of
`Technology
`C. B. Suresh. Royal Melbourne Institute of Technology
`
`100 JAFMAS: A Mnltiagent Application
`Development System
`Deepiira Chauhan. University of Cincinnati
`Albert D. Baker. University of Cincinnati
`
`1 0 8 Agent Design Patterns: Elements of Agent
`Application Design
`Yariv Aridor. IBM Tokyo Research Laboratory
`Danny B. Lange. General Magic Inc.
`
`iii
`
`Exh. p. 3
`
`

`
`Case 1:14-cv-00598-LPS Document 1-3 Filed 05/12/14 Page 58 of 110 PageID #: 76
`Case 1:14—cv-00598 LPS Do
`"
`cument 1-3 F'| d
`_
`Ie 05/12/14 Page 58 of 110 Pa e|D #;
`
`
`
`.‘V
`
`
`
`..‘’.t....-.
`
`.t'
`
`
`t.,._.,._l.._._,“"44.~n-u\'
`-».\u-‘
`
`PAPER SESSION 5: WWW AGENTS I
`Monday. May it
`-.
`
`116 CiteSeer: An Autonomous Web Agent for
`Automatic Retrieval and Identification of
`Interesting Publications
`Kurt D. Bollacker. University ofTexas at Austin and NEC
`Research Institute
`Steve Lawrence. NEC Research Institute
`C. Lee Giles. NEC Research Institute and University of
`Maryland
`
`124 Query Restart Strategies for Web Agents
`Prasad Chalasani. Carnegie Mellon University
`Somesh Jha. Carttegie Mellon University
`Onn Shchory. Carnegie Mellon University
`Katia Sycara. Carnegie Mellon University
`
`13 '2 WebMate: A Personal Agent for Browsing
`and Searching
`Liren Chen. The Robotics Institute, Carnegie Mellon
`University
`Katia Sycata. The Robotics Institute. Carnegie Mellon
`University
`
`PANEL SESSION
`‘Z200 pm - 3:30 pm
`Monday. May it
`Session Chair: Andrew Stern. PF. Magic
`
`1 4 0 AUTONOMOUS AGENTS IN ENTERTAINMENT
`Joe Bates. Carnegie Mellon University
`Bruce Blurnberg. MIT Media Lab
`Athomas Goldberg. NYU Media Research Laboratory
`Barbara Hayes—Roth. Stanford University
`Andrew Stern. PF. Magic
`
`PAPER SESSION 6: AGENTS FOR
`INFORMATION MANAGEMENT I
`Monday. May 11
`4:00 pm - 5:30 pm
`
`141 Concept Features in Re: Agent, an
`Intelligent Email Agent
`Gary Boone. Georgia Institute of Technology
`
`for Collecting Application Usage
`I49 Agents
`Data Over
`the Internet
`, David M. Hilbert . University of California. Irvine
`David F. Redmiles. University of California, Irvine
`
`PAPER SESSION 7: SYNTHETIC AGENTS
`4:00 pm — 5:30 pm
`Monday. May 11
`
`165 A Social-Psychological Model
`Actors
`Daniel Rousseau. Machi.-ta So iens
`Barbara Hayes-Roth. Stanfor University
`
`for Synthetic
`
`in Annotated Worlds
`17 3 Agents
`Patrick Doyle. Stanford University
`Barbara Hayes-Roth. Stanford University
`
`131 Story-tnorpltlng in the Affective Reasoning
`Paradigm: Generating Stories Senti-
`Automatically for Use with "Emotionally
`Intelligent" Multimedia Agents
`Clark Elliott; DePani University
`Jacek Brzezinski, DePatti University
`Sanjay Sheth. DePaul University
`Robert Salvatoriello. DePanl University
`
`PANEL SESSION
`4:00 pm — 5:30 pm
`Monday, May ll
`Session Chair: Hyacinth S. Nwana, BT Labs
`
`1 8 9 AGENT-MEDIATED ELECTRONIC
`COMMERCE-. ISSUES, CHALLENGES AND
`SOME VIEWPOINTS
`Hyacinth S. Nwana. BT Labs
`Jeff Rosenschcin, Hebrew University
`Tuomas Sandholm. Washington University
`Carlos Sierra. IIIA
`Pattie Macs. MIT Media Laboratory
`Rob Guttmann. MIT Media Laboratory
`
`PAPER SESSION 8: MOBILE SOFTWARE
`AGENTS
`Tuesday. May I!
`
`10:30 am - I2:30 pm
`
`I97 Market-Based Resource Control
`Agents
`Jonathan Bredin. Darttnottth College
`David Kotz. Dartmouth College
`Daniela Rus. Dartmouth College
`
`for Mobile
`
`for Enabling Mobile User
`205 Mobile Agents
`Aware Applications
`Akhil Sahai. INRIAJRISA
`Christine Morin. INRIA-IRISA
`
`1 5 '1 Adaptive Information Agents in Distributed
`Textual Environments
`Filippo Menczer. University of California. San Diego
`Richard K. Bclew. University of California. San Diego
`
`212 A Secure Marketplace for Mobile Java
`Agents
`Kay Neuenhofen. Sun Microsystetns
`Matthew Thompson. Sun Microsystems
`
`219 Mobile Agents on the Digital Battlefield
`Martin 0. Hofmann. Lockheed Martin ATL
`Amy McGovern. University of Massachusetts
`Kenneth R. Whitebread. Lockheed Martin ATL
`
`Exh. p. 4
`
`

`
`Case 1:14-cv-00598-LPS Document 1-3 Filed 05/12/14 Page 59 of 110 PageID #: 77
`Case 1:14—cv-00598 LPS
`-
`Document 1-3
`'
`‘A
`Ftled 05/12/14 Page 59 of 110 Page|[) #; 77
`
`PAPER SESSION 11: COMPUTATIONAL
`MARKET SYSTEMS
`Tuesday, May 12
`
`2:00 pm — 3:30 pm
`
`285 A Market Architecture for Multi-Agent
`Contracting
`John Collins. University of Minnesota
`Ben Youttgdahl, University of Minnesota
`Scott Jamison, University of Minnesota
`Bamshad Mobashcr. University of Minnesota
`Maria Gini, University of Minnesota
`
`293
`
`.
`
`'
`
`for Heterogeneous
`
`Competitive Scenarios
`Trading Agents
`Juan A. Rodriguez-Aguilar, IHA, CSIC
`Francisco J. Martin, HIA, CSIC
`Pablo Noriega. HIA. CSIC
`Pete Garcia, HIA. CSIC
`Caries Sierra. IIIA, CSIC
`
`PAPER SESSION 9: LEARNING AND
`ADAPTATION I
`10:30 am - 12:30 pm
`Tuesday‘. May 12
`
`226
`
`A Multiagent Perspective of Parallel and
`Distributed Machine Learning
`Gerhard Weill. Teclmisclte Uttiversitiit Mt'itichelt
`
`Learning Situation-Dependent Costs:
`Improving Planning from Probabilistic
`Robot Execution
`Karen Zita Haigh, Carnegie Mellon University
`Manuela M. Veloso. Carnegie Mellon University
`
`239
`
`Online Learning about Other Agents in a
`Dynamic Multiagent System
`lunling Hu. Uttiversity of Michigan
`Michael P. Wellmatt. University of Michigan
`
`II Et
`
`Is Relevant to the Effects of
`Learning What
`Actions for a Mobile Robot
`Matthew D. Schmill, University of Massachusetts
`Michael T. Rosenstein. University of Massachusetts
`Paul R. Cohen, University of Massachusetts
`Paul Utgoff. University of Massachusetts
`
`PAPER SESSION 10: LIFELIKE AGENTS
`l0:3Ct am - 12:30 pm
`Tuesday, May 12
`
`Behavioural Self-Organlzatieon in Lifelike
`Agents
`liming Lia. Hong Kong Baptist University
`Hong Qtn, Hong Kong Baptist University
`
`Integrating Reactive and Scripted Behaviors
`in a Life-Like Presentation Agent
`Elisabeth Andre, German Research Center for Artificial
`Intelligence (DFKI)
`Research Center for Artificial
`Thomas Rist. German
`Intelligence (DFKI)
`Jochcn Miiller. German Research Center for Artificial
`Intelligence (DFKI)
`BodyChat: Autonomous Communicative '
`Behaviors
`in Avatars
`Hartnes Htigni Vilhjalrnssott, MIT Media Laboratory
`Justine Cassel]. MIT Media Labo
`
`254
`
`261
`
`269
`
`277
`
`“Vision Fulfilled
`Arthur Allen, Cltaris Software Systetrts
`
`301
`
`The Michigan Internet AuctionBot: A
`Configurable Auction Server for Human and
`Software Agents
`Peter R. Wurtnan. University of Michigan
`Michael P. Wellman. University of Michigan
`William E. Walsh, Uttiversity of Michigan
`
`PAPER SESSION 12: AGENT CONTROL
`ARCHITECTURES I
`Tuesday. May 12
`
`2:00 pm - 3:30 pm
`
`3 0 9
`
`Increasing Agent Autonomy in Dynamic
`Environments
`Subrata Das. Cltaries River Analytics
`Alp-er Caglayan, Charles River Analytics
`Paul Gonsalves, Charles River Analytics
`
`Methods
`for Complex,
`Dynamically Simulated Agents: Adonis
`Dances the Macarena
`Maia I. Matatic, University of Sotttltern California
`Victor B. Zotdan. Georgia institute of Technology
`Zachary Mason, Brandeis University
`
`325 Integrating Active Perception with an
`Autonomous Robot Architecture
`Glenn Wasson. University of Virgitta
`David Kortenkamp. Metrica, Inc., TRACLabs
`Eric Huber. Metrica. lnc.. TRACLabs
`
`Exh. p. 5
`
`

`
`Case 1:14-cv-00598-LPS Document 1-3 Filed 05/12/14 Page 60 of 110 PageID #: 78
`t 1-3 Filed 05/12/14 Page 60 of 110 Page|D #: 78
`Case 1:14-cv-00598-LPS Documen
`'
`
`
`
`-c-J"
`
` t-‘zeroV-'t-
`.~.:.fir._
`
`-..-._-3-1"‘‘~l‘,-n.\.
`
`VIDEO SESSION I
`Tuesday. May I2
`
`2200 pm - 3:30 pm
`
`3 32 STEVE: A Pedagogical Agent
`Reality
`Jeff Rickei, University of Southern California
`W. Lewis Johnson. University of Southern California
`
`for Virtual
`
`334
`
`Virtual Petz: A Hybrid Approach to Creating
`Autonomous, Lifelike Dogz and Catz
`Andrew Stem, PF. Magic
`Adam Frank. PF. Magic
`Ben Rcsner. PF. Magic
`
`The Use of Space Discretization for
`Autonomous Virtual Humans
`Stikanth Bandi. Swiss Federal Institute of Technology
`Daniel Thalmann. Swiss Federal Institute of Technology
`
`PAPER SESSION 13: AGENTS FOR
`INFORMATION MANAGEMENT II
`4:00 pm - 5:00 pm
`Tuesday. May I2
`
`338
`
`The ATL Postmaster: A System for Agent
`Collaboration and Information
`Dissemination
`Jennifer Kay. Lockheed Martin Advanced Technology
`Laboratories
`Julius Etzi. Lockheed Martin Advanced Technology
`Laboratories
`Gouthaln Rao. Lockheed Martin Advanced Technology
`Laboratories
`Jon 'I11ies, Lockheed Martin Advanced Technology
`Laboratories
`
`343
`
`Understanding Jokes: A Neural Approach to
`Content-Based Information Retrieval
`Stcphane Zrehen. USC Brain Project
`Michael A. Arbih. USC Brain Project
`
`
`
`iLI
`
`-\'§'
`
`VIDEO SESSION II
`Tuesday. May 12
`
`4:00 pm - 5:00 pm
`
`3 5 0 The RETSINA Multiagent System: Towards
`Integrating Planning, Execution and
`Information Gathering
`Katia Sycara. Carnegie Mellon University
`Anandcep S. Pannu. Carnegie Mellon University
`
`352
`
`Information Systems
`
`Domain-Adaptive
`(DAIS)
`Kenneth R. Wititebtcad, Lockheed Martin Advanced
`Technology Laboratories
`Harley J. Stein. Lockheed Martin Advanced Technology
`Laboratories
`
`PAPER SESSION 14: AGENT CONTROL
`ARCHITECTURES 11
`Wednesday. May ti!
`
`10:30 am - 12:00 pm
`
`354 Action Selection in an Autonomous Agent
`with a Hierarchical Distributed Reactive
`Planning Architecture
`Vincent Decugis. Groupe_d'Etndes Sous-Marine de
`l'Atlantiqne
`Jacques Fetber. Robotique et Microélectronique de
`Montpellier
`
`362 Issues in Temporal Reasoning for
`Autonomous Control Systems
`Nicola lvluscettola. Recon: Technologies. NASA Antes
`Research Center
`Paul Morris. Caelurn Research. NASA Antes Research
`Center
`Barney Pell. Riacs. NASA Antes Research Center
`Ben Smith. Jet Propulsion Laboratory, California
`Institute of Technology
`
`3 6 9 A Hybrid ProcedurallDeductlve Executive for
`Autonomous Spacecraft
`Barney Pell. Riacs, NASA Ames Research Center
`Edward B. Gamble. Jet Propulsion
`Laboratory/California Institute of Technology
`Etann Gat. Jet Propulsion Laboratory/California
`institute of Technology
`Ron Keesing. Caelum Research. NASA Antes Reseach
`Center
`James Kuricn. Caehun Research, NASA Antes Reseach
`Center
`William Millar. Caelum Research, NASA Antes Reseach
`Center
`P. Pandurang Nayak. Riacs. Nasa Atnes Research Center
`Christian Plaunt. Coehnn Research. NASA Ames Research
`Center
`Brian C. Williams. NASA Antes Research Center
`
`PAPER SESSION 15: ENGINEERING AGENT
`SYSTEMS 11
`Wednesday. May 13
`
`10:30 am - 12:00 pm
`
`3'17 Personal Security Agent: KQML-Based PKI
`Qi He. Carnegie Mellon University
`Katia P. Sycara. Carnegie Mellon University
`Timothy W. Finin. University of Maryland Baltimore
`County
`
`385 Pitfalls of Agent-Oriented Development
`Michael Wooldtidge. Queen Mary & Westfield College,
`University of London
`Nicholas R. Jennings, Queen Mary & Westfield College,
`University of London
`
`392 Liability for Autonomous Agent Design
`Carey Heckman. Stanford Law School
`Jacob O. Wobbmck. Stanford Symbolic Systems
`Program
`
`vi
`
`Exh. p. 6
`
`

`
`Case 1:14-cv-00598-LPS Document 1-3 Filed 05/12/14 Page 61 of 110 PageID #: 79
`Case 1'14-cv-00598
`'
`
`—LPS Document 1 3-
`
`.Filed 05/12/14 Page 61 of 110 Pagem #: 79
`
`F:-————_._—.
`
`AGENTS II
`PAPER SESSION 16: WWW l0:30 am -
`l2;90 pm
`Wednesday. May 13
`400 A Web-based Information System that
`Reasons with Structured Collections of Text
`T&T Labs-Research
`
`William W. Cohen. A
`
`.
`
`Daniel Boley, Universi ofMinnesota
`Categorization
`But Hong (Sam) Han. University of Minnesota
`Maria Gini, University ofMinnesota
`Robert Gross, University ofMinnesota
`K le Hastings. University ofMinnesota
`GeorgeKat
`is, University ofMinnesota
`Bamshad Mobashct, University ofMinnesota
`Vipin Kumar, University of Minnesota
`
`416 SHOPPEIUS EYE: Using Locatiombased
`Filtering for a Shopping Agent
`in the
`PilY5i°31 W07”
`
`Consiiiting
`
`Andrew E. Fano, Andersen
`
`1:30 pm - 2:30 pm
`
`oach for Adaptive
`4 22 A History-Based Appr
`Robot Beha or
`in Dynamic Environments
`Francois Michaud. Université de Siierbrooiie
`UIIIVETNT)’ 0f501”i'9"" California
`Maja 3. Matatic,
`0 Effects of Misconception on Reciprocative
`A gents
`Sandip Sen, The University of Tulsa
`Ariish Bisv.-as. The University of Tiiisa
`
`PAPER SESSION 18: A
`Wednesday, May I3
`
`PPLICATIONS
`1:30 pm - 2:30 pm
`
`Non-Intrusive
`436 SANI: A
`for Creating
`Framework and Agent
`Sandeep Chatierjee, Massnciiiisetts iristititte of
`Intelligent
`Interactive Homes
`Technology
`441 Reactive Agents
`for Adaptive Image
`Kong Baptist University
`
`Analysis
`Jiming Lin, Hang
`
`POSTER SESSION
`Wednesday, May I3
`
`1:30 pm - 3:00 pm
`
`al Autonomous Agents with
`449 Artifici
`Artificial Emotions
`Luis Miguel Botelho. ISCTE
`Heider Coelho, University of Lisbon
`for
`451 Mobility and Management Support
`Mobile Agents
`hung-Hsitig University
`Wen-Shyen E. Chen, National C
`S. T. Su,National Chung-Hsing University
`Yao-Nan Lien, National Ciiengciii University
`H. T. Shit. institutefor information Iitdnstry
`HnilingLin. Institutefor Information Industry
`4 53 Dynamic Software Agents
`for Business
`Intelligence Applications
`Qiining Chen, HP Laboratories
`Parvathi Chnndi, HP Laboratories
`Umeshwaf Da,.a1,Hp f_abo,.a;o,.,,_,,
`Meiehun Hsu, HP Laboratories
`455 RohoTA: An Agent Colony Architecture for
`rting Education
`Suppo
`. Forbes, Northwestern University
`Sven E. Kuehne, Nortitwestern University
`
`457 A Formal Treatment of Distributed
`Matchmaking
`'
`iion University
`Sotnesh Jha. Carnegie Me'e Mellon University
`Prasad Chalasani. CarnegiMellon Univers_i'ty
`Orin Shehoiy, Carnegie
`Katie Sycata, Carnegie Mellon University
`
`4 6 3 Einbo
`Sociology
`Simon Penny, Carnegie Mellon University
`
`465 The Mnlti-Agent-Based Schedule Calculator
`(MASC) System
`Arena Sankaranarayanan. Fideiity investinents
`Maia Mataric, University ofSouthern California
`467 ULTIMA RATIO — Should Hamlet Kill
`
`Claudius?
`Michael Schroeder. University OfHt1tttt01'£t‘
`Plewe. Franz-Kiinstier-Str
`Daniela A.
`of Magcieburg
`Andreas Raab, University
`
`vii
`
`Exh. p. 7
`
`

`
`Case 1:14-cv-00598-LPS Document 1-3 Filed 05/12/14 Page 62 of 110 PageID #: 80
`Fase 1:14—cv—OO598—LPS Document 1-3 Filed 05/12/14 Page 62 of 110 Page|D #: 80
`
`469
`
`471
`
`473
`
`475
`
`Combining Mediation and Bidding
`Mechanisms
`for Agent-Based Manufacturing
`Scheduling
`Weiming Shen, The University of Calgary
`Douglas H. Norrie. The University of Calgary
`
`Cognition and Affect: Architectures and
`Tools
`
`Aaron Sloman, University of Birtninghant
`Brian Logan, University of Birmingham
`
`An Open Architecture for Emotion and
`Behavior Control of Autonomous Agents
`Juan D. Velasquez, MIT Artificial Intelligence
`Laboratory
`Masahiro Fajita. Sony Corporation
`Hiroaki Kitano, Sony Computer Science Laboratory, Inc.
`
`The University of Michigan Digital Library
`Service Market Society
`José M. Vidal, University of Michigan
`Tracy Mullen. University of Michigan
`Peter Weinstein, University of Michigan
`Edmund H. Durfee. University ofMichigan
`
`477 Author Index
`
`viii
`
`Exla. p. 3
`
`

`
`Case 1:14-cv-00598-LPS Document 1-3 Filed 05/12/14 Page 63 of 110 PageID #: 81
`Case 1:14—cv—OO598—LPS Document 1-3
`Filed 05/12/14 Page 63 of 110 Page|D #' 81
`
`3---—-~—m~r
`
`; l
`
`Acknowledgments
`We would like to take this opportunity to thank everyone
`involved with the organisation of Agents '98. First, we would
`like to thank Tim Finin and the area chairs, for their hard work
`and first-rate scientific evaluation of papers and videos
`submitted. Dan Weld must be singled out for handling no less
`than l25 papers on software agents — somewhat more than
`anyone anticipated! Maja Mataric and Clark Elliott did
`excellent jobs of handling papers and videos in the robotics
`and synthetic agcntsfagents for entertainment areas. Keith
`Decker handled publicity for the conference, which involved
`(amongst other things) changes to the conference WWW site
`seemingly every hour. Milind Tambc handled the finances for
`the conference, and Maria Gini did a superb job of local
`organization. Mike Huhns handled the organization of
`workshops, an innovation of this year's conference, which
`added significantly to the richness of the conference program.
`Anand Rao was tutorial chair, and Henry Kautz and Robin '
`Murphy handled the demonstration sessions, of software
`agents and robotic agents respectively. Afsanch Haddadi
`managed the poster sessions for the conference, and David
`Musliner
`the exhibits. Bamshad Mobasher handled
`registrations on behalf of
`the conference. The program
`committee did a typically thorough and conscientious job of
`reviewing a very large number of papers.
`We would like to thank the staff of AAAI, and in particular
`Carol Hamilton and Keri Vasser, for handling the submissions
`to the conference, putting together
`the proceedings so
`professionally, and generally giving excellent advice on
`organizational matters of all kinds. From ACM, we would like
`to extend our gratitude to Alisa Rivkirt for her help with the
`proceedings. We would also like to thank ACM S1GART and
`ACM in general for their continued support, which has been of
`enormous benefit
`in so quickly establishing Autonomous
`Agents a major event
`in the conference calendar, and in
`addition our other sponsors, without whose support Agents '98
`would have been significantly less
`interesting. The
`enthusiastic support of so many sponsors is a good indicator of
`how seriously the world is taking agent technology.
`Finally, we would like to extend our thanks to Lewis Johnson
`for his sound and timely advice, and the Autonomous Agents
`steering committee for their helpful suggestions at many
`points throughout the long (and often exhausting) process of
`conference organization.
`
`Katia P. Sycara (Carnegie Mellon University, USA) and
`Michael Wooldridge (Queen Mary & Westfield College, UK)
`March l99B
`
`INTRODUCTION
`Autonomous agents are computer systems that are capable of
`independent action in dynamic, unpredictable environments.
`Agents are also one of the most important and exciting areas of
`research and development in computer science today. Agents
`are currently being applied in domains as diverse as computer
`games and interactive cinema,
`information retrieval and
`filtering, user interface design, and industrial process control.
`The‘ aim of the Agents '93 conference is to bring together
`researchers and developers from industry and academia in order
`to report on the latest scientific and technical advances, discuss
`and debate the major issues, and showcase the latest systems.
`The First International Conference on Autonomous Agents
`(Agents '97) was held in Marina del Rey, California,
`in
`February 1997. It was attended by nearly 500 people, and
`received media coverage from such varied and widely-respcctcd
`organizations as Wired magazine, the New York Times, and
`CNN. It was generally reckoned to have created something of a
`stir far beyond the audience that
`the organizers originally
`expected. All this made Agents '97 a hard act to follow —— but
`we believe that we have succeeded in Agents '98.
`It is only a year since the first Autonomous Agents conference,
`and yet in that time, agent technology has come a long way. At
`Agents '97, delegates were talking about the possibility of
`commercializing agent technology; of using agents in “real"
`systems. In just one year, we might have expected to see a few
`tentative efforts in this direction. But
`to our pleasure and
`surprise, we have seen agent technology adopted not just by a
`few research projects, but by nearly all major players in the
`commercial software marketplace. Agents are now an everyday
`component of software, with agent-enabled features rapidly
`becoming accepted as the norm, rather than as the exception.
`Autonomous Agents '98 is a vivid illustration of the latest
`developments in agent technology. Like its predecessor, it is
`focused around three main strands:
`software
`- Software agents, which are situated in a
`environment, and typically act as “expert assistants" to
`users carrying out some task.
`- Robotic
`agents, which are physically embodied
`autonomous robots, sensing and acting in the everyday
`physical world.
`virtual
`shared
`inhabit
`0 Synthetic
`agents, which
`environments, often in the form of computer games, virtual
`theater, or interactive cinema.
`Nearly 130 technical papers were submitted to the conference,
`and all were rigorously reviewed by the program committee. Of
`these submissions, only 57 were accepted as full
`technical
`papers. This high rejection rate is more a reflection of the care
`and thought that
`the program committee and area chairs put
`into the review and selection process than the standard of
`papers submitted. The overall outcome of the review process is
`a selection of papers, videos, and software and hardware
`demonstrations
`that
`showcase the very best of agent
`technology today.
`We are confident that Agents '93 will confirm the Autonomous
`Agents series of conferences as a key forum for presenting
`work in the applications of agent technology.
`
`ix
`
`Exh. p. 9
`
`

`
`Case 1:14-cv-00598-LPS Document 1-3 Filed 05/12/14 Page 64 of 110 PageID #: 82
`Case 1:14—cv—OO598—LP
`-
`-
`8 Document 1 3 Filed 05/12/14 Page 64 of 110 PageID #: 82
`
`WebMate : A lgersonal Agent for Browsing and Searching*
`
`Liren Chen and Katia Sycara
`
`The Robotics Institute
`
`Carnegie Mellon University
`
`Pittsburgh, PA. 15213
`
`tchen@cs.cmu.edu, l<atia@cs.crnu.edu
`
`i i E
`
`Abstract
`
`The World-Wide Web is developing very fast. Currently,
`finding useful information on the Web is a time consum-
`ing process.
`In this paper, we present WebMate, an agent
`that helps users to effectively browse and search the Web.
`WebMate extends the state of the art in Web-based informa-
`tion retrieval in many ways. First, it uses multiple TF-IDF
`vectors to keep track of user interests in different domains.
`These domains are automatically teamed by WebMate. Sec-
`ond, WebMate uses the Trigger Pair Model to automatically
`extract keywords for refining document search. Third, dur-
`ing search, the user can provide multiple pages as sirnilar—
`itylrelevance guidance for the search. The system extracts
`and combines relevant keywords from these relevant pages
`and uses them for keyword refinement. Using these tech-
`niques, WebMate provides effective browsing and searching
`help and also compiles and sends to users personal newspa-
`per by automatically spiding news sources. We have experi-
`mentally evaluated the performance of the system.
`
`Area: Software Agents
`
`Keywords: Information Agents. Instructability, Knowledge
`acquisition and accumulation, long-terrn adaptation and
`teaming, user modeling
`
`1
`
`Introduction
`
`The Web is full of information and resources. People have at
`least three ways to find information they need: (1) by brows-
`ing (following hyper-links that seem of interest to them), (2)
`by sending a query to a search engine, such as Altavista, (3)
`by following existing categories in search engines, such as
`:__m..__.
`‘This research has been supported in pm by ARPA contract F336l5-93-I-I330.
`and by ONR Grant NGOOI4-96-1222.
`
`P-:rn1i\sion to make digitalthard copies ol'all or part ofthis material for
`personal or claxsrooni use is granted without fee provided that the copies
`are not made or distributed for prolit or commercial advantage. the copy-
`right notice, the title oftlte publication and iLs date appear. and notice is
`gt‘-L-,11[m[¢._,pyi“gislty pennis.‘aiot10l'ACi\L Inc. To copy otltenwse.
`to .-epnh'li,eh_ to post on servers or to rcdistribttte to lists. requires prior
`specific permission andfor fee.
`Autonomous Agents ‘)8 Minneapolis MN USA
`Copyrigltl l99l-l U—E‘)7‘)l-‘J83-lr‘).‘l-‘ 5...‘S5_llU
`
`Yahoo or Lycos. The problem is that people have to spend
`a lot of time and effort to navigate but may not find in-
`teresting personalized information. However, it is difficult
`to find the wanted information because a user can‘t accu-
`rately express what he wants and search engines don’t adapt
`their search strategies according to different users. More-
`over,
`the problem is exacerbatcd because the information
`sources have high “noise”,
`i.e. most of the pages are ir-
`relevant to a particular user’s interests. Intelligent software
`agents are being developed to deal with these issues.
`Intelligent agents are programs that act on behalf of their
`human users to perform laborious information-gathering tasks
`[I] and they are one of the "hot" topics in Information Sys-
`tems R&D at the moment. The last ten years have seen a
`marked interest in agent-oriented technology, spanning ap-
`plications as diverse as information retrieval, user interface
`design and network management.
`in this paper, we present WebMate, a personal software
`agent that accompanies a user when he browses and searches
`and provides intelligent help ‘.
`For clarity of presentation, the WebMate capabilities will
`be presented in roughly two categories: (1) learning user in-
`terests incrementally and with continuous update and auto-
`matically providing documents (e.g. a personalized newspa-
`per) that match the user interests, and (2) helping the user
`refine search so as to increase retrieval of relevant docu-
`ments. in section 2, we describe the architecture of the sys-
`tem.
`'Il1e WebMate acts as a proxy and monitors a user’s
`actions. In section 3. we describe the user profile represen-
`tation and learning algorithm {3, 4]. In addition, we provide
`experimental results of compiling a personal newspaper. In
`section 4, we discuss how to use the Trigger Pairs Model to
`extract relevant words to use as keyword refinements to im-
`prove search. We also present utilizing relevance feedback
`[8] during search to dynamically enhance the search for rel-
`evant documents. Finally, related work and our future work
`are described.
`
`‘The Webhiate system has been operating on Web and has been downloaded by
`more than 600 users since it was published in the middle of September 1997 (15 days
`ago). its URL is http:ll'www.cs.cmu.eduJ"softagentsIw-ebmare.
`
`
`
`‘-t.‘-it...s'.u...uuua$..t..
`
`132
`
`Exh. p. 10
`
`

`
`Case 1:14-cv-00598-LPS Document 1-3 Filed 05/12/14 Page 65 of 110 PageID #: 83
`Case 1:14-cv-00598 LPS Do
`‘
`cume t 1-3
`'
`n
`t=I|ed 05/12/14 Page 65 of 110 Page|[) #; 83
`
`In contrast to other systems that learn a user profile and
`use it statically to determine relevant documents, WebMatc
`learns the user profile incrementally and continuously. When
`a new positive example is known, the system updates the
`profile. In order to save on storage space, the system doesn’t
`keep any of the previous positive example documents.
`It
`only keeps the profile learned from those positive examples.
`In this way, the system will adapt to the user’s evolving and
`recent interests.
`WebMate utilizes TF-IDF method [7] with multiple vec~
`tors representation. The basic idea of the algorithm is to
`represent each document as a vector in a vector space so
`that documents with similar content have similar vectors.
`. Each dimension ofthe vector space represents a word and its
`weight. The values ofthe vector elements for a document are
`calculated as a combination of the statistics term frequency
`TF(w, d) (the number of times word to occurs in document
`d) and document frequency DF(w) (the number of docu-
`ments the word to occurs in at least once). From the doc-
`ument frequency the inverse document frequency IDF(to)
`can be calculated.
`
`I DF (to) = log
`
`
`lDl
`DF(u:)
`
`|D| is the total number of documents. The value til‘) of
`an element in the vector is then calculated as the product
`
`at‘) = TF(w,~, cl) >< IDF(w,-)
`
`We have developed an algorithm for multi TF-IDF vector
`learning. The algorithm follows.
`We assume that a user has at most N domains of interest.
`2 Assume the initial profile set is V, [Vi = 0; the predefined
`number of TF-IDF vectors in the profile set is N. the preset
`number of elements of a vector is M. For each positive ex-
`ample (i.e. an HTML documents that the user has marked “I
`like It”), do‘.
`1. Preprocess: parse HTML page, deleting the stop words
`(or non-informative words) such as “a”, “the”, “is",
`“in", etc, stemming the plural noun to its single form
`and inflexed verb to its original form, extracting the
`words in :itle(<TlTLE>). head1(<1-I1 >), head2(<H2>),
`hea