Vellore Institute OF Technology, BHOPAL
Fundamentals In AI & ML
(Department of Computer Science &
Engineering)
Ankit Shrivastava
Vellore Institute Of Technology (VIT),
Bhopal, India
Module-4
Software Agents and Prolog
2
Architecture for Intelligent Agents:
The task of AI is to design an agent program which
implements the agent function. The structure of an
intelligent agent is a combination of architecture and
agent program. It can be viewed as:
Agent = Architecture + Agent Program
Following are the main three terms involved
in the structure of an AI agent:
Architecture: Architecture is the machinery on which
the agent executes its action. It is essentially a device
with embedded actuators and sensors.
3
Agent Program: The Agent Program performs the
execution of the Agent Function. The execution
happens on top of Agent Architecture and produces
the desired function.
Agent Function: This is a function in which actions are
mapped from a certain percept sequence. Percept
sequence refers to a history of what the intelligent
agent has perceived.
f . P *-> A
4
PEAS Representation:
PEAS stands for:
– Performance
– Environment
– Actuators
– Sensors
Performance Measure: The output which we get from
the agent. All the necessary results that an agent gives
after processing comes under its performance.
5
Environment: All the surrounding things and
conditions of an agent fall in this section. It basically
consists of all the things under which the agents work.
Actuators: The devices, hardware or software through
which the agent performs any actions or processes any
information to produce a result are the actuators of the
agent.
Sensors: The devices through which the agent
observes and perceives its environment are the sensors
of the agent.
6
What is PEAS for Self-driving Cars:
Performance: Safety, time, legal drive, comfort.
Environment: Roads, other cars, pedestrians, road
signs.
Actuators: Steering, accelerator, brake, signal, horn.
Sensors: Camera, sonar, GPS, Speedometer,
7
odometer, accelerometer, engine sensors, keyboard.
Agent Communication:
Communication is necessary in order to allow
collaboration, negotiation, cooperation, etc… between
independent entities. For this purpose, it requires a
well-defined, agreed and commonly understood
semantics.
Agent is an active object with the ability to perceive,
reason and act.
Agent has the ability to communicate. This ability is
part perceptron (the receiving of messages) and part
action (the sending of messages).
Communication can enable the agents to coordinate
their actions and behaviour, resulting in systems that
8 are more coherent.
Why do we need Agent
Communication:
To have a interaction between a machine
and a human we need agent
communication.
9
Multi-agent System:
Computerized system composed of multiple
interacting intelligent agents.
Examples:
Robots, humans or human teams.
Contain combined human-agent teams
Distributed systems
10
Collaborative Agents:
A multi-agent system (MAS) may be seen as a
collection of collaborative agents.
They can communicate and cooperate with other
agents, while keeping their autonomy.
They usually negotiate with their peers to reach
mutually acceptable agreements during cooperative
problem solving.
They have limited learning capabilities.
11
Collaborative Agents: Applications
Provide solutions to physically distributed problems
Disaster in a city (police, firemen, ambulances)
Provide solutions to problems with distributed data
sources
Sensor network monitoring a given area
Provide solutions that need distributed expertise
Health care provision (family doctors, nurses,
specialists, laboratory analysis).
12
Negotiation and Bargaining:
Negotiation:
Negotiation is a process in which a joint decision is
made by two or more parties. The parties first
verbalize contradictory demands and then move
towards agreement by a process of concession-making
or search for new alternatives.
The major features of Negotiation are
The language used by the participating agents.
The protocol followed by the agents as they negotiate.
The decision process that each agent uses to determine
13 its positions, concessions and criteria for agreement.
• Systems and Techniques for Negotiation:
Environment-centered
Agent-centered
• The resultant negotiation mechanism should ideally
have the following attributes:
Efficiency: The agents should not waste resources in
coming to an agreement.
Stability: No agent should have an incentive to
deviate from agreed-upon strategies.
Simplicity: The negotiation mechanism should impose
low computational and bandwidth demands on the
agents.
14
Distribution: The mechanism should not require a
central decision maker.
Symmetry: The mechanism should not be biased
against any agent for arbitrary or inappropriate
reasons.
15
Types of Environment:
Worth-oriented domains
State-oriented domains
Task-oriented domains
• A task-oriented domain is one where agents have a set
of tasks to achieve, all resources needed to achieve the
tasks are available, and the agents can achieve the
tasks without help or interference from each other.
• However, the agents can benefit by sharing some of
the tasks.
16
Example: Internet downloading domain
Each agent is given a list of documents that it must
access over the Internet.
17
Bargaining:
• Bargaining is a simple form of the distributive
negotiation process that is both competitive and
positional. Meaning bargaining doesn’t seek to create
value but instead focuses on negotiators claiming
value.
• Bargaining predominates in one-time negotiations, e.g.
buying or selling a car or house when you don’t intend
on having an ongoing relationship. One side usually
attempts to gain an advantage over another to obtain
the best possible agreement.
18
Cooperative game or Non-cooperative game:
There are two ways to model bilateral negotiations:
using cooperative game or non-cooperative game.
In cooperative games, agreements are enforceable or
binding, and its possible for the agents to negotiate
outcomes that are mutually beneficial.
In non-cooperative game, the agents are self-Interested
and thus they have incentive to deviate from an
agreement to improve his/her utility.
Thus, a same game would have the different outcome
between cooperative games and non-cooperative games.
19
Example:
Silent Confess
Silent 8,8 0,10
Confess 10,0 5,5
Agents can cooperatively find the mutually
most beneficial outcome.
20
Augmentation among agents:
“A verbal and social activity of reason aimed at
increasing (or decreasing) the acceptability of a
controversial standpoint for the listener or reader, by
putting forward a constellation of propositions (i.e.
arguments) intended to justify (or refute) the
standpoint before a rational judge”.
Argument: Reasons / Justifications supporting a
conclusion.
Represented as: support-> conclusion
21
Informational arguments: Beliefs-> Belief
e.g. If it is cloudy, it might rain.
• Motivational arguments: Beliefs, Desires-> Desire
e.g. If it is cloudy and you own a raincoat then put the
raincoat.
• Practical arguments: Belief, Sub-Goals-> Goal
e.g. If it is cloudy and you own a raincoat then put the
raincoat.
• Social arguments: Social commitments-> Goal,
Desire
e.g. I will stop at the corner because the law say so.
e.g. I can’t do that, I promise to my mother that won’t.
22
Cut and Fail Predicate:
Fail Predicate-
Fail predicate simply fails the rule.
The fail forces backtracking in an attempt to unify with
another clause.
// Code snippet
a(x):- b(x), c(x), fail.
a(x):- d(x).
b(1).
b(4).
c(1).
c(3).
d(4). Query: ?- a(x).
23
Cut predicate-
Cut Predicate remove alternatives then forbids values that
otherwise would be ‘returned’ by means of X binding.
Cut Predicate always succeeds.
Fail predicate always fail.
Symbol of cut predicate is “!”
//code snippet
a(x):- b(x), ! c(x).
a(x):- d(x).
b(1).
b(4).
c(1).
c(3).
24 d(4).
Difference Between Cut and Fail
Predicate:
a(x):- b(x),!, c(x), fail. a(x):- b(x), c(x), fail, !.
a(x):- d(x). a(x):- d(x).
b(1). b(1).
b(4). b(4).
c(1). c(1).
c(3). c(3).
d(4). d(4).
Query: Query:
? – a(x).
?- a(x).
FALSE
X=4
25
Thank you
26
