COMP 4640 Intelligent and Interactive Systems Intelligent Agents

1
COMP 4640Intelligent and Interactive
SystemsIntelligent Agents

• Chapter 2

2
Rational Agents

• Rational Agent For each possible percept
  sequence, a rational agent should select an
  action that is expected to maximize its
  performance measure, given the evidence provided
  by the percept sequence and whatever built-in
  knowledge the agent has.

3
Rational agents

• Rationality is distinct from omniscience
  (all-knowing with infinite knowledge)
  
• Agents can perform actions in order to modify
  future percepts so as to obtain useful
  information (information gathering, exploration)
  
• An agent is autonomous if its behavior is
  determined by its own experience (with ability to
  learn and adapt)
  

4
PEAS

• PEAS Performance measure, Environment,
  Actuators, Sensors
• Must first specify the setting for intelligent
  agent design
  
• Consider, e.g., the task of designing an
  automated taxi driver
  
• Performance measure
• Environment
• Actuators
• Sensors

5
PEAS

• Must first specify the setting for intelligent
  agent design
  
• Consider, e.g., the task of designing an
  automated taxi driver
  
• Performance measure Safe, fast, legal,
  comfortable trip, maximize profits
  
• Environment Roads, other traffic, pedestrians,
  customers
  
• Actuators Steering wheel, accelerator, brake,
  signal, horn
  
• Sensors Cameras, sonar, speedometer, GPS,
  odometer, engine sensors, keyboard
  

6
PEAS

• Agent Medical diagnosis system
• Performance measure Healthy patient, minimize
  costs, lawsuits
• Environment Patient, hospital, staff
• Actuators Screen display (questions, tests,
  diagnoses, treatments, referrals)
  
• Sensors Keyboard (entry of symptoms, findings,
  patient's answers)

7
PEAS

• Agent Part-picking robot
• Performance measure Percentage of parts in
  correct bins
• Environment Conveyor belt with parts, bins
• Actuators Jointed arm and hand
• Sensors Camera, joint angle sensors

8
PEAS

• Agent Interactive English tutor
• Performance measure Maximize student's score on
  test
• Environment Set of students
• Actuators Screen display (exercises,
  suggestions, corrections)
• Sensors Keyboard

9
Structure of Intelligent Agents

• The objective of AI is the design and application
  of agent programs that implement mappings between
  percepts and actions
• An agent can be viewed as a program that is
  developed to run on a particular architecture
  (some computing device).
• The architecture
• makes percepts available,
• runs the agent program,
• sends actions to the effectors

10
Types of Agent Programs

• Intelligent systems are typically composed of a
  number of intelligent agents.
• Each agent interacts (directly or indirectly)
  with one or more aspects of an environment.
• This type of agent interaction is similar to
  what we see in sports, business, and other
  organizations that are composed of a number of
  different agents with different responsibilities
  working together for the common good.

11
Environment types

• There are a number of different agent programs
  however, many can be classified as one of the
  following
• Agent Environments
• Fully vs. Partially Observable (Accessible vs.
  inaccessible)
• Deterministic vs. Stochastic (non-deterministic)
• Episodic vs. Sequential (non-episodic)
• Static vs. dynamic
• Discrete vs. continuous

12
Environment types

• Fully observable (vs. partially observable) An
  agent's sensors give it access to the complete
  state of the environment at each point in time.
  
• Deterministic (vs. stochastic) The next state of
  the environment is completely determined by the
  current state and the action executed by the
  agent. (If the environment is deterministic
  except for the actions of other agents, then the
  environment is strategic)
• Episodic (vs. sequential) The agent's experience
  is divided into atomic "episodes" (each episode
  consists of the agent perceiving and then
  performing a single action), and the choice of
  action in each episode depends only on the
  episode itself.

13
Environment types

• Static (vs. dynamic) The environment is
  unchanged while an agent is deliberating. (The
  environment is semidynamic if the environment
  itself does not change with the passage of time
  but the agent's performance score does)
• Discrete (vs. continuous) A limited number of
  distinct, clearly defined percepts and actions.
  
• Single agent (vs. multiagent) An agent operating
  by itself in an environment.
  

14
Environment types

• Chess with Chess without Taxi driving
• a clock a clock
• Fully observable Yes Yes No
• Deterministic Strategic Strategic No
• Episodic No No No
• Static Semi Yes No
• Discrete Yes Yes No
• Single agent No No No
• The environment type largely determines the agent
  design
  
• The real world is (of course) partially
  observable, stochastic, sequential, dynamic,
  continuous, multi-agent
  

15
Agent functions and programs

• An agent is completely specified by the agent
  function mapping percept sequences to actions
• One agent function (or a small equivalence class)
  is rational
  
• Aim find a way to implement the rational agent
  function concisely
  

16
Table-lookup agent

• \inputalgorithms/table-agent-algorithm
• Drawbacks
• Huge table
• Take a long time to build the table
• No autonomy
• Even with learning, need a long time to learn the
  table entries

17
Agent types

• Four basic types in order of increasing
  generality
  
• Simple reflex agents
• Model-based reflex agents
• Goal-based agents
• Utility-based agents

18
Simple reflex agents
19
COMP-4640 Intelligent Interactive
SystemsSimple Reflex Agent I
Performance Measure Get to the bowl Percept
Sequence (x,y) coordinates Agents Knowledge of
the Environment No Knowledge Set of
Actions 0 1 2 7 c 3 6 5 4
Rule Base R01 If at(5,0) ? action(2) R02 If
at(6,1) ? action(2) R03 If at(7,2) ?
action(2) R04 If at(8,3) ? action(1) R05 If
at(8,4) ? action(1) R06 If at(8,5) ? action(1)
R07 If at(8,6) ? action(0) R08 If at(7,7) ?
action(0) R09 If at(6,8) ? action(0) R10 If
at(5,9) ? action(0)
20
COMP-4640 Intelligent Interactive
SystemsSimple Reflex Agent II
Performance Measure Get to the bowl Percept
Sequence (x,y) coordinates Agents Knowledge of
the Environment No Knowledge Set of
Actions 0 1 2 7 c 3 6 5 4, MTB (Move Towards
Bowl)
21
COMP-4640 Intelligent Interactive
SystemsSimple Reflex Agent II
Rule Base R01 If MTB(x) ? clear(x) ? action(x)
? remove(lastMove(y)) ? assert(lastMove(x)) R02
If MTB(x) ? ?clear(x) ? assert(escape_phase) R03
If MTB(x) ? clear(x) ? escape_phase ? action(x) ?
remove(lastMove(y))? remove(escape_phase) ?
assert(lastMove(x)) R04 If escape_phase ?
?lastMove(5) ? clear(1) ? action(1) ?
retract(lastMove(_)) ? assert(lastMove(1)) R05
If escape_phase ? ?lastMove(7) ? clear(3) ?
action(3) ? retract(lastMove(_)) ?
assert(lastMove(3)) R06 If escape_phase ?
?lastMove(1) ? clear(5) ? action(5) ?
retract(lastMove(_)) ? assert(lastMove(5)) R07
If escape_phase ? ?lastMove(3) ? clear(7) ?
action(7) ? retract(lastMove(_)) ?
assert(lastMove(7))
22
COMP-4640 Intelligent Interactive
SystemsSimple Reflex Agent II
Rule Base R01 If MTB(x) ? clear(x) ? action(x)
? remove(lastMove(y)) ? assert(lastMove(x)) R02
If MTB(x) ? ?clear(x) ? assert(escape_phase) R03
If MTB(x) ? clear(x) ? escape_phase ? action(x) ?
remove(lastMove(y))? remove(escape_phase) ?
assert(lastMove(x)) R04 If escape_phase ?
?lastMove(5) ? clear(1) ? action(1) ?
retract(lastMove(_)) ? assert(lastMove(1)) R05
If escape_phase ? ?lastMove(7) ? clear(3) ?
action(3) ? retract(lastMove(_)) ?
assert(lastMove(3)) R06 If escape_phase ?
?lastMove(1) ? clear(5) ? action(5) ?
retract(lastMove(_)) ? assert(lastMove(5)) R07
If escape_phase ? ?lastMove(3) ? clear(7) ?
action(7) ? retract(lastMove(_)) ?
assert(lastMove(7))
23
Model-based reflex agents
24
Goal-based agents

25
COMP-4640 Intelligent Interactive
SystemsGoal-Based Agent
Performance Measure Get to the Agents
Knowledge bowl Percept Sequence (x,y)
coordinates Agents Knowledge of the
Environment No Knowledge Set of Actions 0 1
2 7 c 3 6 5 4, GeneratePath
Rule Base R01 If ?path_ok ? GeneratePath ?
assert(path_ok) R02 If path_ok ? getMove(x,y,a)
? clear(a) ? action(a) R03 If path_ok ?
getMove(x,y,a) ? ?clear(a) ? retract(path_ok) ?
retract(getMove(c,d,e))
26
Utility-based agents
27
COMP-4640 Intelligent Interactive
SystemsUtility-Based Agent
Performance Measure Get to the bowl Using the
Shortest Path Percept Sequence (x,y)
coordinates Agents Knowledge of the
Environment No Knowledge Set of Actions 0 1
2 7 c 3 6 5 4, GeneratePath
Rule Base R01 If ?path_ok ? GeneratePath(UF) ?
assert(path_ok) R02 If path_ok ? getMove(x,y,a)
? clear(a) ? action(a) R03 If path_ok ?
getMove(x,y,a) ? ?clear(a) ? retract(path_ok) ?
retract(getMove(c,d,e))
28
Learning agents
29
Learning Agent Examples

• Interactive Animated Pedagogical
  AgentsMicrosoft agentWhy people hate
  clippy?Intelligent tutoring systems Agents on
  Websites with characters to guide
  usersmySimon.combuy.comextempo.comananova.com
  Ikea.com

30
Collaborative Agents
0
31
Collaborative Agents
5000
32
Collaborative Agents
10000
33
Collaborative Agents
0
34
Collaborative Agents
500
35
Collaborative Agents
3000
36
Collaborative Agent Examples

• Collaborative Agents
• CMU
• Advanced Mechantronics Lab

37
More Agent Examples

• Virtual Humans Conversational Agents
• Conversational Agent Applications
• Virtual Patient Audio
• Virtual Pediatric Patient Video
• Video Just Talk
• Sample Animations Video