|||

courses

Quick search

  • Georgia Tech OMSCS
    • Georgia Tech Resources
    • Computability, Complexity & Algorithms
    • Computer Networking
    • Knowledge Based AI - CS7376-O01
    • Software Architecture & Design
    • Database Systems Concepts and Design
    • Artificial Intelligence
      • Game Playing
      • Search
      • Pre-processing for Search
      • Search Examples
      • Constraint Satisfaction
      • Simulated Annealing
      • Probability
      • Reference
      • Bayes Nets
      • Machine Learning
      • Lesson 1 Challenge #7: Expectimax 2
      • Short search questions
      • Searching Rubik’s Cube
      • Lesson 3 Challenge 2
      • Problem 1
      • Problem 2
      • Problem 3
      • Bayes Net Calculation
      • Numpy References
      • Logic and Planning
      • Planning under uncertainty
      • Pattern Recognition Through Time
      • AI Resources
      • Math for AI
    • Machine Learning
    • Compilers: Theory and Practice
    • Computer Vision
    • Computational Photography
    • Artificial Intelligence for Robotics
    • Introduction to Operating Systems
    • Practice Tests
    • Software Analysis and Testing
  • Coursera Courses
  • Courses in EDX
  • CodeSchool Notes
  • Udemy
  • Kubernetes
  • Python
  • Go Programming Language
  • Trading
  • Resources
  • Ideas for Improvement

Search¶

  • http://norvig.com/

Challenge 1 Tri City Search¶

https://dl.dropbox.com/s/9hrrmfgnc7sc1ix/Screenshot%202017-03-11%2016.50.57.png?dl=0
  • Tri-directional A* search

Challenge 2 Rubik’s Cube¶

https://dl.dropbox.com/s/f3e3lapwf16ii40/Screenshot%202017-03-11%2016.52.48.png?dl=0 https://dl.dropbox.com/s/7k6rxfsnrgr9syx/Screenshot%202017-03-11%2016.53.15.png?dl=0
  • Finding Optimal Solutions to Rubik’s Cube Using Pattern Databases

  • God’s Number is 26 in the Quarter-Turn Metric

Problem Solving¶

https://dl.dropbox.com/s/cw98aqwstjw1mrj/Screenshot%202017-03-11%2016.54.27.png?dl=0
  • Complexity from the many choices.

https://dl.dropbox.com/s/fktehlsp8b76hpe/Screenshot%202017-03-11%2016.55.08.png?dl=0
  • Complexity comes from partial observability.

Question: Can we find the direction from Arad to Bucharest from this map?

https://dl.dropbox.com/s/xioo8g9rmnbmulu/Screenshot%202017-03-11%2016.56.53.png?dl=0

What is a problem¶

Agent is given a problem of coming up with a sequence of actions to find a path from Arad to Bucharest.

https://dl.dropbox.com/s/9tov18zruhnbxh5/Screenshot%202017-03-11%2016.58.37.png?dl=0

Definition of the problem

  • Initial State

  • Actions(s) -> {a1, a2, a3 }

  • Result(s, a) -> s’

  • GoalTest(s) -> T|F

  • Path Cost

https://dl.dropbox.com/s/z8pjwwntc678y6u/Screenshot%202017-03-11%2017.09.26.png?dl=0

Example Route Finding¶

https://dl.dropbox.com/s/vwfcrvjjwfwicoy/Screenshot%202017-03-11%2017.13.28.png?dl=0

Tree Search¶

https://dl.dropbox.com/s/jaygycapiax8t1f/Screenshot%202017-03-11%2017.15.33.png?dl=0

Breadth-First Search

https://dl.dropbox.com/s/gq6xr3yxvcnwm15/Screenshot%202017-03-11%2017.17.54.png?dl=0

Tree Search Continued¶

https://dl.dropbox.com/s/zi6hnouw0x3hf8g/Screenshot%202017-03-11%2017.19.15.png?dl=0 https://dl.dropbox.com/s/74uvaa3krzyh7qb/Screenshot%202017-03-11%2017.20.29.png?dl=0 https://dl.dropbox.com/s/v8kdyacspkdaxr9/Screenshot%202017-03-11%2017.20.57.png?dl=0

In the preliminary algorithm, A is repeated since we are not keeping track of explored states. Ideally, we would not add duplicates from backtracking.

Graph Search¶

https://dl.dropbox.com/s/7hp7h5hhp81eoj4/Screenshot%202017-03-11%2017.22.19.png?dl=0

Breadth First Search-1¶

https://dl.dropbox.com/s/ra52b797qflj2qw/Screenshot%202017-03-11%2017.23.22.png?dl=0

Breadth First Search 3¶

https://dl.dropbox.com/s/k838t1nbn743xsj/Screenshot%202017-03-11%2017.28.05.png?dl=0

Uniform Cost Search¶

  • Cheapest path cost search.

https://dl.dropbox.com/s/b4454ghdkkpmn3g/Screenshot%202017-03-11%2017.31.07.png?dl=0 https://dl.dropbox.com/s/zqog7y1l32nk1l9/Screenshot%202017-03-11%2017.32.21.png?dl=0 https://dl.dropbox.com/s/bynvzvkyrgjxwyg/Screenshot%202017-03-11%2017.33.42.png?dl=0 https://dl.dropbox.com/s/5vtnf4tq4zvcabe/Screenshot%202017-03-11%2017.34.19.png?dl=0 https://dl.dropbox.com/s/s4nhgdrqk96pu8o/Screenshot%202017-03-11%2017.35.51.png?dl=0 https://dl.dropbox.com/s/lokqughck4vcnqe/Screenshot%202017-03-11%2017.38.24.png?dl=0

Search Comparison¶

https://dl.dropbox.com/s/1ynre2ilbo651w2/Screenshot%202017-03-11%2017.39.24.png?dl=0 https://dl.dropbox.com/s/bvrgjux6xkm6l0s/Screenshot%202017-03-11%2017.40.51.png?dl=0

Search Comparison - 1¶

https://dl.dropbox.com/s/tmkg51idgy7sggk/Screenshot%202017-03-11%2017.42.45.png?dl=0

Search Comparison 2¶

https://dl.dropbox.com/s/d0akq8tmrt3badd/Screenshot%202017-03-11%2017.44.16.png?dl=0

Are these algorithms complete?

https://dl.dropbox.com/s/zy6043vs3ig1b42/Screenshot%202017-03-11%2017.45.13.png?dl=0

More On Uniform Cost¶

https://dl.dropbox.com/s/yza7askvs5usxao/Screenshot%202017-03-11%2017.47.28.png?dl=0

Greedy Best-First Search

https://dl.dropbox.com/s/aua84jir6v725uu/Screenshot%202017-03-11%2017.49.11.png?dl=0 https://dl.dropbox.com/s/pqldcllvcv93uok/Screenshot%202017-03-11%2017.50.12.png?dl=0

A* = Greedy Best First Search + Uniform Cost Search

A* Search¶

https://dl.dropbox.com/s/jg6ijzqml33r9mi/Screenshot%202017-03-11%2017.53.16.png?dl=0 https://dl.dropbox.com/s/2me6ldgipmc8wem/Screenshot%202017-03-11%2017.53.50.png?dl=0 https://dl.dropbox.com/s/mpnflh4bu1vnv01/Screenshot%202017-03-11%2017.56.02.png?dl=0 https://dl.dropbox.com/s/hk4r585m7m6wsg7/Screenshot%202017-03-11%2017.57.19.png?dl=0 https://dl.dropbox.com/s/l1r7e0ubl7o5uru/Screenshot%202017-03-11%2017.58.18.png?dl=0 https://dl.dropbox.com/s/npc2k23bs1gwnsn/Screenshot%202017-03-11%2017.59.21.png?dl=0 https://dl.dropbox.com/s/a4onhk9xayz0nv5/Screenshot%202017-03-11%2018.00.59.png?dl=0 https://dl.dropbox.com/s/b7bn4nwlomzbi4v/Screenshot%202017-03-11%2018.01.38.png?dl=0

Admissible Heuristic Function¶

Heuristic function is admissible if h(s) <= true cost, rather than necessarily being strictly smaller than the true cost.

  • h should never overestimate the distance to the goal.

  • h should be optimistic

  • h is admissible.

Optimistic Heuristic¶

https://dl.dropbox.com/s/mpxiux0z5x8jvkk/Screenshot%202017-03-11%2018.03.51.png?dl=0

State Spaces¶

https://dl.dropbox.com/s/l53hhz0iclp8e8x/Screenshot%202017-03-11%2018.05.24.png?dl=0 https://dl.dropbox.com/s/5vnahasoimx6bfb/Screenshot%202017-03-11%2018.06.11.png?dl=0
  • Robot can be A or B = 2

World itself

  • State A can be dirty or not = 2

  • State B can be dirty or not = 2

Total = 2 x 2 x 2 = 8 state spaces.

State Spaces 2¶

https://dl.dropbox.com/s/u2oocwpyrzv27kp/Screenshot%202017-03-11%2018.10.18.png?dl=0 https://dl.dropbox.com/s/9aj4zpi7vryjngn/Screenshot%202017-03-11%2018.11.31.png?dl=0

Sliding Blocks Puzzle¶

https://dl.dropbox.com/s/disppsbj849av7u/Screenshot%202017-03-11%2018.13.41.png?dl=0

Sliding Blocks Puzzle 2¶

https://dl.dropbox.com/s/zpx12n437kjy5fb/Screenshot%202017-03-11%2018.16.57.png?dl=0
  • Generating a relaxed problem.

Problems With Search¶

https://dl.dropbox.com/s/6xv9y25454x7in3/Screenshot%202017-03-11%2018.19.42.png?dl=0

A Note On Implementation¶

https://dl.dropbox.com/s/zkdn86iy1tfsejy/Screenshot%202017-03-11%2018.22.05.png?dl=0

References¶

  • Korf, 1997, Finding Optimal Solutions to Rubik’s Cube Using Pattern Databases.

  • Goldberg, 2011. Reach for A* An Efficient Point-to-Point Shortest Path Algorithm

  • Goldberg & Harrelson, March 2003. Computing the Shortest Path A* Search Meets Graph Theory.

  • Gutman, 2004. Reach-based Routing A New Approach to Shortest Path Algorithms Optimized for Road Networks.

<Page contents

>Page contents:

  • Search
    • Challenge 1 Tri City Search
    • Challenge 2 Rubik’s Cube
    • Problem Solving
    • What is a problem
    • Example Route Finding
    • Tree Search
    • Tree Search Continued
    • Graph Search
    • Breadth First Search-1
    • Breadth First Search 3
    • Uniform Cost Search
    • Search Comparison
    • Search Comparison - 1
    • Search Comparison 2
    • More On Uniform Cost
    • A* Search
    • Admissible Heuristic Function
    • Optimistic Heuristic
    • State Spaces
    • State Spaces 2
    • Sliding Blocks Puzzle
    • Sliding Blocks Puzzle 2
    • Problems With Search
    • A Note On Implementation
    • References
<Game Playing
Pre-processing for Search>
© Copyright 2023, Senthil Kumaran. Created using Sphinx 7.1.2.

Styled using the Piccolo Theme