Lecture 7 Agent Communication and Knowledge Representation Languages

1
Lecture 7 Agent Communication and Knowledge
Representation Languages

• Amund Tveit
• Department of Computer and Information Science
• Norwegian University of Science and Technology
• amund.tveit_at_idi.ntnu.no
• http//www.idi.ntnu.no/amundt/
• 47 4 162-6572

2
Lecture Outline

• Brief KQML Repetition
• KQML Agent Naming
• Facilitators
• Routers
• Other ACLs (e.g. FIPA ACL)
• Conclusions of ACLs
• Knowledge Representation and Ontologies
• Knowledge Interchange Format
• OIL
• Conclusions of the lecture

3
Motivation for Agent Communication (R)

• Language is a very difficult thing to put into
  words Voltaire
• Communication is required for cooperation between
  agents
• Societies can perform tasks no individual agent
  can
• Autonomy encourages disregard for other agents
  internal structure
• Communicating agents need only know a common
  language
• Supports heterogenous agents

4
Terminology of Agent Communication (R)

• Speech Act Formal subset of natural language
  representing actions
• ACL Agent Communication Language
• FIPA - Foundation for Intelligent Physical Agents
• KQML Knowledge Query and Manipulation Language
• Ontology World Model, formal description of
  agent domain
• Encoding syntactic representation of ACL
  messages
• ACC Agent Communication Channel (FIPA)
• KSE Knowledge Sharing Effort
• KIF Knowledge Interchange Format
• Router entity that sends incoming messages in
  correct direction
• Facilitator An agent that includes a Router

5
KQML Statement Structure (R)

• KQML Statements consists of
• A performative
• Parameters and context information
• General syntax
• (KQML-performative
• sender word
• receiver word
• language word
• ontology word
• content expression
• ...)

6
KQML full example (R)

• (ask-one content (geoloc TRD (?long ?lat))
• ontology geo-model3)
• (ask-all
• content price(IBM, ?price, ?time)
• receiver stock-server
• language standard-prolog
• ontology NYSE-TICS)

7
KQML Agent naming

• System for mapping agents into names is important
  in most ACLs
• KQML assumes that names are local
• A can register with B under the name Alice
• A can register with C under the name Albert
• Doesnt preclude the use of a central Agent Name
  Server, an architecture used by most systems
• What gets registered under a name? Contact
  information like
• name(amundbot, tcpip, cavenan.idi.ntnu.no,80).
• name(amundbot,smtp,amundbot_at_jfipa.org)
• name(amundbot,http,www.jfipa.org80/)

8
Facilitators - I

• Agents performing various (useful) communication
  services
• Maintain registry of service names
• Forward messages to named services
• Route messages based on content
• Provide matchmaking
• Provide mediation and translation services

9
Facilitators - I
10
Facilitators - II
11
Facilitators III ( Car rental ex.)
12
Facilitators IV
13
Facilitators - V
14
Routers

• Content-independent message routers
• Each KQML-agent is associated with its own
  separate router process
• Router handles all outgoing/incoming ACL messages
• Outgoing messages can specify a particular agent
  address (to agent or router)
• Message can specify a description of context
• Delivery of messages is not guaranteed

15
Other ACLs

• Telescript supports mobile computing
• FIPA ACL competing/extending KQML
• FIPA vs KQML
• Both are based on speech act
• Different set of performatives
• FIPA has a more formal basis
• FIPA can describe interaction protocols
• FIPA is probably becoming the standard ACL

16
What is FIPA?

• The Foundation for Intelligent Physical Agents
• non-profit organization.
• FIPAs purpose is to promote
• agent-based applications,
• Services
• equipment.
• FIPAs goal is pursued by
• Creating internationally agreed specifications
• Specifications maximises interoperability
  between
• agent-based applications,
• services and
• equipment.

17
Who is FIPA?

• FIPA operates through the open international
  collaboration of member organisations, which are
  companies and universities active in the agent
  field.
• Companies
• Alcatel, Boeing, British Telecom, Deutsche
  Telekom, France Telecom, Fujitsu, Hitatchi, HP,
  IBM, Fujitsu, Hewlett Packard, IBM, Intel,
  Lucent, NEC, NHK, NTT, Nortel, Siemens, SUN,
  Telia, Toshiba, etc.
• Universities and Research Institutes
• GMD, EPFL, Imperial, IRST, etc.
• Government Agencies
• DARPA

18
FIPA Encodings and package

• Fipa supports three main types of syntax
  encoding
• Extensible Markup Language XML
• Proprietary Binary format
• Lisp-like format
• Separates between Envelopes and Messages

19
Example - XML-encoded FIPA ACL

• lt?xml version"1.0"?gt
• ltenvelopegt
•   ltparams index"1"gt
•     lttogt
•       ltagent-identifiergt
•         ltnamegtreceiver_at_foo.comlt/namegt
•         ltaddressesgt
•           lturlgthttp//foo.com/acclt/urlgt
•         lt/addressesgt
•       lt/agent-identifiergt
•     lt/togt
•     ltfromgt
•       ltagent-identifiergt
•         ltnamegtsender_at_bar.comlt/namegt
•         ltaddressesgt
•           lturlgthttp//bar.com/acclt/urlgt
•         lt/addressesgt
•       lt/agent-identifiergt
•     lt/fromgt

20
Binary encoded FIPA XML

• 0xfe 0x00 0x97 0x12 0x20 0x31 0x11 0x06 0x19 0x15
  0x37 0x62 0x59 0x20 0x02 0x03 0x02
• r  e  c  e  i  v  e  r  _at_  f 
  o  o  .  c  o  m  0x00
• 0x02 h  t  t  p    /  /  f  o 
  o  .  c  o  m  /  a 
• c  c  0x00 0x01 0x01 0x02 s  e  n  d 
  e  r  _at_  b  a  r  . 
• c  o  m  0x00 0x02 h  t  t  p   
  /  /  b  a  r  .  c 
• o  m  /  a  c  c  0x00 0x01 0x01 0x08
  n  o    e  n  c  r 
• y  p  t  i  o  n  0x00 0x0a h  t 
  t  p    /  /  b  a 
• r  .  c  o  m  /  a  c  c  0x00
  0x20 0x31 0x11 0x06 0x19 0x15 0x37
• 0x62 0x59 0x20 0x03 1  2  3  4  5  6 
  7  8  9  0x00 0x01
•  

21
Shortcomings of current ACLs

• Intentional level description which mental
  attitudes, what definitions?
• Problems with mental attitudes from theory to
  practice
• Can all desirable communication primitives be
  modeled after speech acts? Should they?
• Flexible description of agents capabilities and
  advertising of such capabilities.
• How can we test an agents compliance with the
  ACL?
• Ease of extending an ACL

22
Conclusion of current ACLs

• Agent Communication Languages have a common basis
  speech act
• Syntax is well specified, but current research is
  on describing semantics

23
Motivation for KIF
Language is a very difficult thing to put into
words Voltaire

• Creating language for development of intelligent
  applications
• Creating language for common interchange format
• Note
• KIF is not intended for interaction with humans
• KIF is not intended to be internal representation
  for knowledge within computer programs

24
KIF Features

• Prefix version of 1st order predicate calculus
  (logic)
• Declarative semantics
• Logically comprehensive
• Provides representation of knowledge about
  representation of knowledge
• Additional features
• Translatability
• Readability
• Usability

25
KIF General Overview

• Variables
• ?singlevar, _at_sequence
• Operators
• Term Operators listof, if, ..
• Sentence Operators not, and, /, ...
• Rule Operators gtgt, ..
• Definition Operators defobject, deffunction
• Constants
• object, function, relational, logical
• Expressions
• Word or a finite sequence of words (i.e.
  Sentences)

26
KIF examples - I

• Terms
• (size chip1)
• ( (sin theta)(cos theta))
• (if (gt theta 0) theta (-theta))
• (setofall ?x (above ?x chip1))
• Sentences
• (Prime 565762761)
• (gt (sin theta) (cos phi))
• (not (gt (sin theta) 0)))
• (gt ( (width chip1) (length chip2))((width
  chip1)(length chip2)))
• (gt (gt ?x 0) (positive ?x))

27
KIF Examples - II

• Definitions
• (defrelation leq(?x ?y) (not (gt ?x ?y)))
• (defrelation natural(?x) (and (integer ?x) (gt
  ?x 0)))
• Procedures
• (progn (fresh_line t)(print Hi!)(fresh_line t))
• Meta-Knowledge
• (believes John (material moon jarlsberg))
• (gt (believes John ?p) (believes Mary ?p))
• Rules
• (ltlt (flies ?x)(bird ?x))

28
Exercise!

• How would you describe the following utterances
  in KIF
• A cat is on the mat
• Every cat is on a mat
• A person is between a rock and a hard place
• John is going to Boston by bus
• Tom believes Mary wants to marry a sailor

29
Agent Vocabularies

• Agents should use words consistently
• Open-ended Dictionary
• Each word has
• English description for understanding by humans
• KIF (or alternative) for understanding by
  computers
• Dictionary would contain multiple ontologies for
  many areas

30
KQML and KIF example

• (tell
• sender Agent1
• receiver Agent2
• language KIF
• ontology Blocks-World
• content (AND (Block A)(Block B)(On A B))

31
Semantics on the Web - terminology

• Semantic Web
• RDF Resource Decription Framework
• OIL Open Inference Layer
• XML/XML Schema
• Ontolingua
• OKBC Open Knowledge Base Connectivity
• DAML DARPA Agent Markup Language
• DAML-S Service description language based on
  DAML
• UDDI Universal Description, Discovery and
  Integration
• WSDL Web Services Description Language
• SHOE Simple HTML Ontology Extensions
• ebXML e-business XML

32
Ontology Inference Layer - OIL

• OIL is used to specify and exchange ontologies
• Three roots of OIL
• Description Logics
• Frame-based Systems
• Web-standards XML and RDF
• Small, not too expressive (as possibly the case
  with Ontolingua)

33
OIL Example

• ontology-container
• title african animals
• subject animal, food, vegetarians
• description example ontology describing african
  animals
• Language en-uk
• Source http//www.url.com/
• ontology-definitions
• slot-def eats
• inverse is-eaten-by
• class-def animal
• class-def plant

34
Exercise!

• How would you create an ontology in OIL for this
  scenario
• Alumni (post-student) database

35
OIL References

• http//www.cse.lehigh.edu/heflin/courses/sw-fall0
  1/