02 - Intelligent Agents Handout
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View 02 - Intelligent Agents Handout as PDF for free.
More details
- Words: 853
- Pages: 4
Objectives Intelligent Agents
You should !
!
Dr. Richard J. Povinelli
Copyright Richard J. Povinelli
!
Page 1
rev 1.1, 8/28/2003
Agents and environments
Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
Page 2
Vacuum-cleaner world
Agents include humans, robots, softbots, thermostats, …. The agent function maps from percept histories to actions: !
be able to provide a definition of a rational agent. be able to compare and contrast various agents including reflex, goal-based, and utility-based agents. be able to classify the environment in which a particular agent operates.
Percepts: location and contents, e.g., [A,Dirty] Actions: Left, Right, Suck, NoOp
f : P* -> A
The agent program runs on the physical architecture to produce f Copyright Richard J. Povinelli
Page 3
rev 1.1, 8/28/2003
A vacuum-cleaner agent Percept sequence [A,Clean] [A,Dirty] [B,Clean] [B,Dirty] [A,Clean], [A,Clean] [A,Clean], [A,Dirty]
Copyright Richard J. Povinelli
Fixed performance measure evaluates the environment sequence
Action
Right Suck Left Suck Right Suck
! !
!
What is the right function? Can it be implemented in a small agent program?
Copyright Richard J. Povinelli
one point per square cleaned up in time T? one point per clean square per time step, minus one per move? penalize for > k dirty squares?
A rational agent chooses whichever action maximizes the expected value of the performance measure given the percept sequence to date Rational ≠ omniscient Rational ≠ clairvoyant Rational ≠ successful Rational ⇒ exploration, learning, autonomy
if status = Dirty then return Suck else if location = A then return Right else if location = B then return Left
rev 1.1, 8/28/2003
Page 4
Rationality
function Reflex-Vacuum-Agent([location, status]) returns an action
Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
Page 5
Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
rev 1.1, 8/28/2003
Page 6
Page 1
CAT – PEAS for an Automatic Taxi
PEAS To design a rational agent, we must specify the task environment Consider, e.g., the task of designing an automated taxi: Performance measure !
Describe the ! ! ! !
What we measure an agents against.
For an automatic taxi. Work with a partner for 5 minutes.
Environment !
Preformance measures Environment Actuators Sensors
The world in which the agent operates.
Actuators !
How the agent can modify its environment
Sensors !
How the agent can sense its environment.
Copyright Richard J. Povinelli
Page 7
rev 1.1, 8/28/2003
PEAS for an Automatic Taxi
Copyright Richard J. Povinelli
Agent functions and programs
Performance measures
An agent is completely specified by the agent function mapping percept sequences to actions In principle, one can supply each possible sequence to see what it does. Obviously, a lookup table would usually be immense. One agent function (or a small equivalence class) is rational Aim: find a way to implement the rational agent function concisely An agent program takes a single percept as input, keeps internal state:
!
Environment !
Actuators !
Sensors
function SKELETON-AGENT(percept) returns action static: memory, the agent’s memory of the world memory ← UPDATE-MEMORY (memory, percept) action ← CHOOSE-BEST-ACTION (memory) memory ← UPDATE-MEMORY (memory, action) return action
!
AIMA Slides © Stuart Russell and Peter Norvig, 1998 Chapter 2 Copyright Richard J. Povinelli
Page 8
rev 1.1, 8/28/2003
Page 9
rev 1.1, 8/28/2003
Agent types
AIMA Slides © Stuart Russell and Peter Norvig, 1998 Chapter 2 Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
Page 10
Simple Reflex Agent
Four basic types in order of increasing generality simple reflex agents reflex agents with state goal-based agents utility-based agents
AIMA Slides © Stuart Russell and Peter Norvig, 1998 Chapter 2 Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
Copyright Richard J. Povinelli
Page 11
Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
rev 1.1, 8/28/2003
Page 12
Page 2
Reflex Agents with State
Copyright Richard J. Povinelli
Goal-based Agents
Page 13
rev 1.1, 8/28/2003
Utility-based Agents
Copyright Richard J. Povinelli
Page 14
rev 1.1, 8/28/2003
Environment Types Accessible vs. hidden Deterministic vs. stochastic Episodic vs. nonepisodic Static vs. dynamic Discrete vs. continuous
Copyright Richard J. Povinelli
Page 15
rev 1.1, 8/28/2003
CAT – Environment Types
Copyright Richard J. Povinelli
Environment Types Example
With a partner fill in the following table. You have 5 minutes Solitaire
Backgammon
Solitaire
Internet shopping
Page 16
rev 1.1, 8/28/2003
Taxi
Backgammon
Internet shopping
Taxi
Observable?? Deterministic??
Accessible?? Deterministic??
Episodic??
Episodic??
Static??
Static?? Discrete??
Discrete??
Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
Copyright Richard J. Povinelli
Page 17
Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
rev 1.1, 8/28/2003
Page 18
Page 3
AlMA code The code for each topic is divided into four directories: ! !
!
!
!
agents: code defining agent types and programs algorithms: code for the methods used by the agent programs environments: code defining environment types, simulations domains: problem types and instances for input to algorithms Often run algorithms on domains rather than agents in environments.
AIMA Slides © Stuart Russell and Peter Norvig, 1998 Chapter 2 Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
Copyright Richard J. Povinelli
Page 19
rev 1.1, 8/28/2003
Page 4
You should !
!
Dr. Richard J. Povinelli
Copyright Richard J. Povinelli
!
Page 1
rev 1.1, 8/28/2003
Agents and environments
Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
Page 2
Vacuum-cleaner world
Agents include humans, robots, softbots, thermostats, …. The agent function maps from percept histories to actions: !
be able to provide a definition of a rational agent. be able to compare and contrast various agents including reflex, goal-based, and utility-based agents. be able to classify the environment in which a particular agent operates.
Percepts: location and contents, e.g., [A,Dirty] Actions: Left, Right, Suck, NoOp
f : P* -> A
The agent program runs on the physical architecture to produce f Copyright Richard J. Povinelli
Page 3
rev 1.1, 8/28/2003
A vacuum-cleaner agent Percept sequence [A,Clean] [A,Dirty] [B,Clean] [B,Dirty] [A,Clean], [A,Clean] [A,Clean], [A,Dirty]
Copyright Richard J. Povinelli
Fixed performance measure evaluates the environment sequence
Action
Right Suck Left Suck Right Suck
! !
!
What is the right function? Can it be implemented in a small agent program?
Copyright Richard J. Povinelli
one point per square cleaned up in time T? one point per clean square per time step, minus one per move? penalize for > k dirty squares?
A rational agent chooses whichever action maximizes the expected value of the performance measure given the percept sequence to date Rational ≠ omniscient Rational ≠ clairvoyant Rational ≠ successful Rational ⇒ exploration, learning, autonomy
if status = Dirty then return Suck else if location = A then return Right else if location = B then return Left
rev 1.1, 8/28/2003
Page 4
Rationality
function Reflex-Vacuum-Agent([location, status]) returns an action
Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
Page 5
Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
rev 1.1, 8/28/2003
Page 6
Page 1
CAT – PEAS for an Automatic Taxi
PEAS To design a rational agent, we must specify the task environment Consider, e.g., the task of designing an automated taxi: Performance measure !
Describe the ! ! ! !
What we measure an agents against.
For an automatic taxi. Work with a partner for 5 minutes.
Environment !
Preformance measures Environment Actuators Sensors
The world in which the agent operates.
Actuators !
How the agent can modify its environment
Sensors !
How the agent can sense its environment.
Copyright Richard J. Povinelli
Page 7
rev 1.1, 8/28/2003
PEAS for an Automatic Taxi
Copyright Richard J. Povinelli
Agent functions and programs
Performance measures
An agent is completely specified by the agent function mapping percept sequences to actions In principle, one can supply each possible sequence to see what it does. Obviously, a lookup table would usually be immense. One agent function (or a small equivalence class) is rational Aim: find a way to implement the rational agent function concisely An agent program takes a single percept as input, keeps internal state:
!
Environment !
Actuators !
Sensors
function SKELETON-AGENT(percept) returns action static: memory, the agent’s memory of the world memory ← UPDATE-MEMORY (memory, percept) action ← CHOOSE-BEST-ACTION (memory) memory ← UPDATE-MEMORY (memory, action) return action
!
AIMA Slides © Stuart Russell and Peter Norvig, 1998 Chapter 2 Copyright Richard J. Povinelli
Page 8
rev 1.1, 8/28/2003
Page 9
rev 1.1, 8/28/2003
Agent types
AIMA Slides © Stuart Russell and Peter Norvig, 1998 Chapter 2 Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
Page 10
Simple Reflex Agent
Four basic types in order of increasing generality simple reflex agents reflex agents with state goal-based agents utility-based agents
AIMA Slides © Stuart Russell and Peter Norvig, 1998 Chapter 2 Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
Copyright Richard J. Povinelli
Page 11
Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
rev 1.1, 8/28/2003
Page 12
Page 2
Reflex Agents with State
Copyright Richard J. Povinelli
Goal-based Agents
Page 13
rev 1.1, 8/28/2003
Utility-based Agents
Copyright Richard J. Povinelli
Page 14
rev 1.1, 8/28/2003
Environment Types Accessible vs. hidden Deterministic vs. stochastic Episodic vs. nonepisodic Static vs. dynamic Discrete vs. continuous
Copyright Richard J. Povinelli
Page 15
rev 1.1, 8/28/2003
CAT – Environment Types
Copyright Richard J. Povinelli
Environment Types Example
With a partner fill in the following table. You have 5 minutes Solitaire
Backgammon
Solitaire
Internet shopping
Page 16
rev 1.1, 8/28/2003
Taxi
Backgammon
Internet shopping
Taxi
Observable?? Deterministic??
Accessible?? Deterministic??
Episodic??
Episodic??
Static??
Static?? Discrete??
Discrete??
Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
Copyright Richard J. Povinelli
Page 17
Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
rev 1.1, 8/28/2003
Page 18
Page 3
AlMA code The code for each topic is divided into four directories: ! !
!
!
!
agents: code defining agent types and programs algorithms: code for the methods used by the agent programs environments: code defining environment types, simulations domains: problem types and instances for input to algorithms Often run algorithms on domains rather than agents in environments.
AIMA Slides © Stuart Russell and Peter Norvig, 1998 Chapter 2 Copyright Richard J. Povinelli
rev 1.1, 8/28/2003
Copyright Richard J. Povinelli
Page 19
rev 1.1, 8/28/2003
Page 4
Related Documents
02 - Intelligent Agents Handout
May 2020 11
Intelligent Agents
May 2020 11
Intelligent Agents
April 2020 14
Intelligent Agents 2
May 2020 7
Intelligent Agents Kb
May 2020 6