AI Spring 2010 Lecture 7

AI Spring 2010 - Artificial Intelligence Lecture 7 Adversarial Search(Part II Chapter 5 Spring edit Click to2010 Master subtitle style Instructor

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon
Click to edit Master subtitle style Artificial Intelligence Lecture 7: Adversarial Search Spring 2010 Instructor: Paul S. Rosenbloom
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
22 Upcoming Events £ Project 1 will be assigned February 4 l Due February 23 by 11:59 pm PST l A program in C++ plus some short answers £ Midterm 1 coming up on February 16 l Covers through Chapter 8 (first order logic) l Minus Chapter 6 (constraint satisfaction problems) and Section 4.2 (local search in continuous spaces)
Background image of page 2
33 Today’s Lecture £ What are games? £ Optimal decisions in games £ l -] search/pruning £ Games of imperfect information l Cutoffs and evaluation functions £ Games that include an element of chance
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
44 Games £ Games are a form of multiagent environment u It matters what other agents do and how this affects success l Even when physical environment is deterministic, other agents inject a form of partially modelable bounded nondeterminism into problems £ Cooperative vs. competitive vs. mixed multiagent environments l Cooperative: identical goals and/or objective function l Competitive: inverse goals and/or objective function l Mixed: cooperative, competitive and/or neutral aspects £ Competitive multiagent environments give rise to adversarial search, a.k.a. games
Background image of page 4
55 Why Study Games? £ Fun; historically entertaining £ Easy to define but difficult to play well l Good play often feasible via search algorithms plus knowledge £ A significant component of the history/success of AI l We are presently in transition from human to machine dominance l Focus here is mostly on traditional board games l Video games require other things, such as virtual humans £ One aspect of multiagent systems
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
66 Games and Search £ What we have seen so far is single-agent search l Solution is usually a single path to a goal state l Focus is frequently on finding optimal solutions l f(n) is an estimate of cost from start to goal through given node l Examples: path planning, scheduling, puzzles £ Games are two-or-more person adversarial search l Solution is usually a branching (tree) structure of possibilities and responses (akin to contingency plans ) u Strict time limits usually force approximate solutions l f(n) : estimate of advantage in game position l Examples: chess, checkers/draughts, othello/reversi, backgammon, go, bridge, poker, monopoly, scrabble
Background image of page 6
77 Types of Games chess, checkers, go, othello, … backgammon, monopoly, … Above, w/ cutoffs Bridge, poker, scrabble, war, Perfect Information Imperfect Information Stochastic Deterministic
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
88 Game Formulation £ Two players: MAX and MIN £ MAX moves first and then take turns until game is over l Each move by a player is called a ply l Winner gets reward, loser gets penalty £ Formulating the game/problem l Initial state: Initial board configuration l Successor function: (move,state) pairs specifying legal moves l Goal/terminal test: Is the game finished? l
Background image of page 8
Image of page 9
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 03/05/2010 for the course CS 561 taught by Professor Moradi during the Spring '09 term at USC.

Page1 / 44

AI Spring 2010 - Artificial Intelligence Lecture 7 Adversarial Search(Part II Chapter 5 Spring edit Click to2010 Master subtitle style Instructor

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online