Professional Documents
Culture Documents
Intelligent agents
By:
Boreshban boreshban@gmail.com
1 / 25
outline:
• Intelligent agents
• Structure of intelligent agents
• Rational agents
• Environment types
• Agent types
2 / 25
Agents
• An agent is any thing that can be viewed as
• Sensors: perceive environment
• Action: act upon environment
3 / 25
Samples of agents
• Human agent: eyes, ears, and other organs for sensors; hands, legs,
mouth, and other body parts for effectors.
4 / 25
Rational agent
• "do the right thing "based on the perception history and the actions
it can perform.
5 / 25
Primary Design Notes (PAGE)
• Perceptions
• Actions
• Goals
• Environments
6 / 25
PAGE Samples
7 / 25
PAGE Samples
Agent: Medical diagnosis system
8 / 25
PAGE Samples
Agent: Part picking robot
9 / 25
PAGE Samples
Agent: Interactive English tutor
• Perceptions: Keyboard
• Actions: Screen display (exercises, suggestions, corrections)
• Goal: Maximize student's score on test
• Environment: Set of students
10 / 25
Autonomy
11 / 25
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.
▫ Example: Chess vs. Taxi driver
12 / 25
Environment types
• 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.
▫ Episodic environments are much simpler because the agent does not
need to think ahead.
13 / 25
Environment types
• Discrete(vs. continuous): A limited number of distinct, clearly
defined percepts and actions.
▫ Example: Chess vs. Taxi driver
14 / 25
Environment types
15 / 25
Agent Program Types
• Look Up Table
• Simple Reflexive
• Goal-based agents
• Utility-based agents
16 / 25
Look Up Table Agents
• Benefits:
▫ Easy to implement
• 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 / 25
Simple Reflex Agents
18 / 25
Simple Reflex Agents
• No Memory, No planning
19 / 25
Reflex agents with states
20 / 25
Reflex agents with states
• no longer-term planning
21 / 25
Goal-based agents
22 / 25
Goal-based agents
23 / 25
Utility-based agents
24 / 25
Utility-based agents
25 / 25