What is a Pattern Database? Pattern Databases. Success Story #1. Success Story #2. Robert Holte Computing Science Dept. University of Alberta
|
|
- Arlene Roberts
- 5 years ago
- Views:
Transcription
1 What is a Pattern Database Pattern Databases Robert Holte Computing Science Dept. University of Alberta November, 00 PDB = a heuristic stored as a lookup table Invented by Culberson and Schaeffer (99) created by abstracting the state space Key properties: guaranteed to be a lower bound guaranteed to be consistent the bigger the better (as a general rule) Success Story # Joe Culberson & Jonathan Schaeffer (99). 5-puzzle (0 states). hand-crafted patterns ( fringe (FR) and corner (CO)) Each PDB contains >500 million entries Used symmetries to compress and enhance the use of the PDBs Used in conjunction with Manhattan Distance (MD) Reduction in size of search tree: MD = * max(md,fr) MD = 7 * max(md,co) MD = 0 * max(md, dovetail(fr,co)) + tricks Success Story # Rich Korf (997) Rubik s Cube (0 9 states). hand-crafted patterns, all used together (max) Each PDB contains over million entries took hour to build all the PDBs Results: First time random instances had been solved optimally Hardest (solution length ) took 7 days Best known MD-like heuristic would have taken a century
2 Example: -puzzle Patterns created by domain abstraction ,0 states original state corresponding pattern Domain = blank 5 7 Domain = blank 5 7 Abstract = blank This abstraction produces 9 patterns Pattern Space Pattern Database goal pattern Pattern Distance to goal 0 Pattern Distance to goal
3 Calculating h(s) Domain Abstraction Given a state in the original problem 7 5 Compute the corresponding pattern Look up the abstract distance-to-goal Domain = blank 5 7 Abstract = blank 7 0,0 patterns Fundamental Questions -puzzle: A* vs. PDB size How to invent effective heuristics Create a simplified version of your problem. Use the exact distances in the simplified version as heuristic estimates in the original. How to use memory to speed up search Precompute all distances-to-goal in the simplified version of the problem and store them in a lookup table (pattern database). # nodes expanded (A*) pattern database size (# of abstract states)
4 Automatic Creation of Domain Abstractions Easy to enumerate all possible domain abstractions Domain = blank 5 7 Abstract = blank They form a lattice, e.g. Domain = blank 5 7 Abstract = blank is more abstract than the domain abstraction above Efficiency Time for the preprocessing to create a PDB is usually negligible compared to the time to solve one problem-instance with no heuristic. Memory is the limiting factor. Making the Best Use of Memory Compression Results Compress an individual Pattern Database Lossless compression Lossy compression must maintain admissibility Allows you to use a PDB bigger than will fit in memory use multiple PDBs instead of just one Merge two PDBs into one the same size Culberson & Schaeffer s dovetailing Jonathan s new idea -disk -peg TOH, PDB based on disks No compression: 5Megs memory,. secs lossless compression: 5k memory,. secs Lossy compression: 9Megs, 5.9 secs 5-puzzle, additive PDB triple (7-7-) No compression: 57Megs memory, 0.09 secs Lossy compression, two PDB triples 57Megs memory, 0.0 secs
5 Max ing Multiple Heuristics Question Given heuristics h and h define h(s) = max ( h(s), h(s) ) Preserves key properties: lower bound consistency Given a fixed amount of memory, M, which gives the best heuristic pattern database (PDB) of size M max ing PDBs of size M/ max ing PDBs of size M/ etc. large pattern database half-size pattern databases s ϕ s ϕ ϕ h(s) h (s) h (s) max 5
6 Many small pattern databases ϕ h (s) s ϕ n h n (s) Rubik s Cube PDB Size n,05,00 7,70,00,,00 Nodes Generated,5,9,9,99,09,99 5,,00 5,9,9 0,,00,5,5 max Summary Rubik s Cube CPU Time RATIO = State Space (x)-puzzle 9-pancake (,)-Topspin ( ops) (,)-Topspin ( ops) (x)-puzzle Best n Ratio Rubik s Cube. 5-puzzle (additive) 5. -puzzle (additive). to 5. #nodes generated using one PDB of size M #nodes generated using n PDBs of size M/n #PDBs Nodes Ratio Time Ratio time/node is.7x higher using six PDBs
7 Early Stopping Techniques for Reducing the Overhead of Multiple PDB lookup IDA* depth bound = 7 g(s) = Stop doing PDB lookups as soon as h > is found. Might result in extra IDA* iterations PDB (s) = 5 next bound is PDB (s) = 7 next bound is 0 Consistency-based Bounding Experimental Results A PDB (A) = PDB (A) = 7 5-puzzle, five additive PDBs (7-7-) Naïve: 0.5 secs Early Stopping: 0.0 secs B Because of consistency: PDB (B) PDB (B) No need to consult PDB Rubik s Cube, six non-additive PDBs Naïve: 7.5 secs Early Stopping:.955 secs Early Stopping and Bounding:. secs 7
8 Static Distribution of Heuristic Values Why Does Max ing Speed Up Search max of 5 small PDBs. large PDB..x nodes generated Runtime Distribution of Heuristic Values Saving Space If h and h are stored as pattern databases, max(h(s),h(s)) requires twice as much space as just one of them. How can we get the benefits of max without using any extra space dovetail two PDBs use smaller PDBs to define max
9 Dovetailing Dovetailing based on the blank Given PDBs for a state space construct a hybrid containing some entries from each of them, so that the total number of entries is the same as in one of the originals. PDB PDB PDB PDB PDB PDB PDB PDB PDB PDB PDB PDB The hope: almost as good as max, but only half the memory. PDB PDB PDB PDB PDB PDB Any colouring is possible Dovetailing selection rule Dovetailing requires a rule that maps each state, s, to one of the PDBs. Use that PDB to compute h(s). Any rule will work, but they won t all give the same performance. Intuitively, strict alternation between PDBs expected to be almost as good as max. 9
10 Dovetailing compared to Max ing Experimental Results Culberson & Schaeffer 99: Dovetailing two PDBs reduced #nodes generated by a factor of.5 compared to using either PDB alone max failures 7 dovetail failures Holte & Newton (unpublished): Dovetailing halved #nodes generated on average Example of Max Failing Depth Bound h h max(h,h) How to generalize Dovetailing to any abstractions of any space,05, 7,99 99,79,,05 9 5,0,5 0,97,9 0 TOTAL 5,5,, 0
11 Example Multiple Lookups in One Pattern Database 5 distance state goal Domain = blank 5 7 Abstract = blank Standard PDB lookup Second lookup, same PDB abstract state abstract goal abstract state abstract goal
12 Relevance Why is this lookup Two Key Properties () Distances are Symmetric relevant to the original state 5 7 () Distances are tile-independent Experimental Results -disk, -peg TOH, PDB of disks Normal: 7. secs Only the second lookup:. secs Both lookups:. secs Additive Pattern Databases 5-puzzle, additive PDB (-7) Normal: 0.0 secs Only the second lookup: 0.07 secs Both lookups: 0.0 secs
13 Adding instead of Max ing Manhattan Distance Heuristic Under some circumstances it is possible to add the values from two PDBs instead of just max ing them and still have an admissible heuristic. For a sliding-tile puzzle, Manhattan Distance looks at each tile individually, counts how many moves it is away from its goal position, and adds up these numbers. This is advantageous because h (s) + h (s) max( h (s), h (s) ) MD(s) = + + = 5 goal state s M.D. as Additive PDBs () In General ϕ (x)= x if x = blank otherwise Partition the tiles in groups, G, G, G k PDB [ϕ (s) ] = ϕ i (x)= x blank if x G i otherwise ϕ (goal) ϕ (s) MD(s) = PDB [ϕ (s) ] + PDB [ϕ (s) ] + PDB [ϕ (s) ]
14 Korf & Felner s Method What s the Difference Partition the tiles in groups, G, G, G k ϕ i (x)= x if x G i blank if x = blank Moves of otherwise cost zero the blank cannot reach this position without disturbing tile or tile. On-demand distance calculation Hierarchical Search To build a PDB you must calculate all abstract distances-to-goal. Only a tiny fraction of them are needed to solve any individual problem. If you only intend to use the PDB to solve a few problems, calculate PDB entries only as you need them. Hierarchical Search
15 Calculate Distance by Searching at the Abstract Level Replace this line: by h(s) = PDB[ φ(s) ] h(s) = search(φ(s), φ(goal) ) (recursive) call to a search algorithm to compute abstract distance to goal for state s Hierarchical Search Abstract space, φ (φ (S)) φ Abstract space, φ (S) φ Original space, S 5-puzzle Results () 5-puzzle Results () Felner s 7-7- additive PDB: takes 0 minutes to build (,00 secs) Solves problems in 0.05 secs (on average) Felner s -7 additive PDB Takes 7 hours to build (5,00 secs) Solves problems in 0.0 secs Hierarchical IDA*, Gigabyte limit Using the same abstraction for all problems, solving takes secs (on average), or 07 secs if the cache is not cleared between problems Max ing over Corner & Fringe abstractions, solving takes 50 secs (on average) Using a customized abstraction for each problem, solving takes 7 secs (on average) 5
16 Thesis topics abound! General Dovetailing A Partial-Order on Domain Abstractions Lattice of domain abstractions Easy to enumerate all possible domain abstractions Domain = blank 5 7 Abstract = blank and to define a partial-order on them, e.g. Domain = blank 5 7 Abstract = blank is more abstract than the domain abstraction above.
17 The LCA of Abstractions General Dovetailing Domain = blank 5 7 Abstract = blank Domain = blank 5 7 Abstract = blank blank 5 7 blank LCA = least-abstract common abstraction Given PDB and PDB defined by ϕ and ϕ Find a common abstraction ϕ of ϕ and ϕ Because it is a common abstraction there exist ϕ and ϕ such that ϕ ϕ = ϕ ϕ = ϕ For every pattern, p, defined by ϕ, set SELECT[p] = ϕ or ϕ Keep every entry (p k,h) from PDB i for which SELECT[ϕ i (p k )]=i. Given state s, lookup SELECT[ϕ(s)](s) 7
Outline. Where Do Heuristics Come From? Part 3. Question. Max ing Multiple Heuristics
Outline Where Do Heuristics Come From? Part Robert C. Holte Computing Science Department University of Alberta 00, Robert Holte Heuristics/Holte Part, Slide Part : Introduction Part : Details Part : Pattern
More informationThe Big Idea. Where Do Heuristics Come From? Puzzles. Applications. (Using Abstraction to Speed Up Search)
Where Do Heuristics Come From? (Using Abstraction to Speed Up Search) Robert C. Holte Computing Science Department University of Alberta The Big Idea Create a simplified version of your problem. Use the
More informationCompressing Pattern Databases
Compressing Pattern Databases Ariel Felner and Ram Meshulam Computer Science Department Bar-Ilan University Ramat-Gan, Israel 92500 Email: felner,meshulr1 @cs.biu.ac.il Robert C. Holte Computing Science
More informationMultiple Pattern Databases
Multiple Pattern Databases R. C. Holte and Jack Newton University of Alberta Computing Science Department Edmonton, Alberta, T6G 2E8, Canada Email: {holte,newton}@cs.ualberta.ca A. Felner and R. Meshulam
More informationMaximizing over multiple pattern databases speeds up heuristic search
Artificial Intelligence 170 (2006) 1123 1136 www.elsevier.com/locate/artint Maximizing over multiple pattern databases speeds up heuristic search Robert C. Holte a,, Ariel Felner b, Jack Newton a, Ram
More informationHeuristic Evaluation Function. 11. Evaluation. Issues. Admissible Evaluations. Jonathan Schaeffer
Heuristic Evaluation Function. Evaluation Jonathan Schaeffer jonathan@cs.ualberta.ca www.cs.ualberta.ca/~jonathan Most of the magic in a single-agent searcher is in the evaluation function To obtain an
More informationLecture 5 Heuristics. Last Time: A* Search
CSE 473 Lecture 5 Heuristics CSE AI Faculty Last Time: A* Search Use an evaluation function f(n) for node n. f(n) = estimated total cost of path thru n to goal f(n) = g(n) + h(n) g(n) = cost so far to
More informationInconsistent Heuristics
Uzi Zahavi Computer Science Bar-Ilan University Ramat-Gan, Israel 925 zahaviu@cs.biu.ac.il Inconsistent Heuristics Ariel Felner Jonathan Schaeffer and Nathan Sturtevant Information Systems Engineering
More informationCompressed Pattern Databases
Journal of Artificial Intelligence Research 30 (2007) 213-247 Submitted 12/06; published 10/07 Compressed Pattern Databases Ariel Felner Department of Information Systems Engineering, Ben-Gurion University
More informationAbstraction-Based Heuristics with True Distance Computations
Abstraction-Based Heuristics with True Distance Computations Ariel Felner and Max Barer Information Systems Engineering Deutsche Telekom Labs Ben-Gurion University Be er-sheva, Israel 85104 felner@bgu.ac.il
More informationInformed search algorithms. Chapter 3 (Based on Slides by Stuart Russell, Dan Klein, Richard Korf, Subbarao Kambhampati, and UW-AI faculty)
Informed search algorithms Chapter 3 (Based on Slides by Stuart Russell, Dan Klein, Richard Korf, Subbarao Kambhampati, and UW-AI faculty) Intuition, like the rays of the sun, acts only in an inflexibly
More informationAdditive Pattern Database Heuristics
Journal of Artificial Intelligence Research 22 (2004) 279-318 Submitted 04/04; published 11/04 Additive Pattern Database Heuristics Ariel Felner felner@bgu.ac.il Department of Information Systems Engineering,
More informationRecent Progress in Heuristic Search: A Case Study of the Four-Peg Towers of Hanoi Problem
Recent Progress in Heuristic Search: A Case Study of the Four-Peg Towers of Hanoi Problem Richard E. Korf Computer Science Department University of California, Los Angeles Los Angeles, CA 90095 korf@cs.ucla.edu
More informationPredicting the Performance of IDA* using Conditional Distributions
Journal of Artificial Intelligence Research 37 (2010) 41-83 Submitted 07/09; published 02/10 Predicting the Performance of IDA* using Conditional Distributions Uzi Zahavi Computer Science Department Bar-Ilan
More informationPartial Pattern Databases
Partial Pattern Databases Kenneth Anderson, Robert Holte, and Jonathan Schaeffer University of Alberta, Computing Science Department, Edmonton, Alberta, T6G 2E8, Canada {anderson, holte, jonathan}@cs.ualberta.ca
More informationLossy Compression of Pattern Databases Using Acyclic Random Hypergraphs
Lossy Compression of Pattern Databases Using Acyclic Random Hypergraphs Mehdi Sadeqi and Howard J. Hamilton Department of Computer Science, University of Regina, Canada {sadeqi2m,hamilton}@cs.uregina.ca
More informationExploiting the Rubik s Cube 12-edge PDB by Combining Partial Pattern Databases and Bloom Filters
Exploiting the Rubik s Cube 12-edge PDB by Combining Partial Pattern Databases and Bloom Filters Nathan R. Sturtevant Dept. of Computer Science University of Denver Denver, CO, USA sturtevant@cs.du.edu
More informationMoving On. 10. Single-agent Search. Applications. Why Alpha-Beta First?
Moving On 10. Single-agent Search Jonathan Schaeffer jonathan@cs.ualberta.ca www.cs.ualberta.ca/~jonathan Two-player adversary search is nice, but not all interesting problems can be mapped to games Large
More informationFinding Optimal Solutions to Sokoban Using Instance Dependent Pattern Databases
Proceedings of the Sixth International Symposium on Combinatorial Search Finding Optimal Solutions to Sokoban Using Instance Dependent Pattern Databases André Grahl Pereira and Marcus Rolf Peter Ritt and
More informationSolving Sokoban Optimally using Pattern Databases for Deadlock Detection
Solving Sokoban Optimally using Pattern Databases for Deadlock Detection André G. Pereira, Marcus Ritt, Luciana S. Buriol Institute of Informatics Federal University of Rio Grande do Sul, Brazil {agpereira,marcus.ritt,buriol
More informationAdditive Pattern Database Heuristics
Additive Pattern Database Heuristics Ariel Felner Ben-Gurion University of the Negev, Beer-Sheva, Israel Richard E.Korf, University of California,Los Angeles Sarit Hanan Bar-Ilan University, Ramat-Gan,
More informationRecursive Best-First Search with Bounded Overhead
Recursive Best-First Search with Bounded Overhead Matthew Hatem and Scott Kiesel and Wheeler Ruml with support from NSF grant IIS-1150068 Wheeler Ruml (UNH) Recursive Best-First Search with Bounded Overhead
More informationLearning from Multiple Heuristics
Mehdi Samadi Computing Science University of Alberta Edmonton, Alberta, Canada T6G E8 msamadi@cs.ualberta.ca Learning from Multiple Heuristics Ariel Felner Information Systems Engineering Dept. Deutsche
More informationUniversity of Alberta. Levi Henrique Santana de Lelis. Doctor of Philosophy. Department of Computing Science
University of Alberta CLUSTER-AND-CONQUER: A PARADIGM FOR SOLVING STATE-SPACE PROBLEMS by Levi Henrique Santana de Lelis A thesis submitted to the Faculty of Graduate Studies and Research in partial fulfillment
More informationPredicting the Size of IDA* s Search Tree
Predicting the Size of IDA* s Search Tree Levi H. S. Lelis a, Sandra Zilles b, Robert C. Holte a a Computing Science Department, University of Alberta, Edmonton, AB, Canada T6G 2E8 b Department of Computer
More informationMinimizing Disk I/O in Two-Bit Breadth-First Search
Proceedings of the Twenty-Third AAAI Conference on Artificial Intelligence (2008) Minimizing Disk I/O in Two-Bit Breadth-First Search Richard E. Korf Computer Science Department University of California,
More informationParallel Algorithms. Single Agent Search COMP 3705/4705. Parallel Search - Simple approaches. Dovetailing
Parallel Algorithms Single Agent Search COMP 3705/4705 Prof. Nathan Sturtevant Lecture 15 For many years it looked like our problems would be solved by increases in processor speed Moore s law predicts
More informationRecent Results from Analyzing the Performance of Heuristic Search
Recent Results from Analyzing the Performance of Heuristic Search Teresa M. Breyer and Richard E. Korf Computer Science Department University of California, Los Angeles Los Angeles, CA 90095 {treyer,korf}@cs.ucla.edu
More informationEstimating Search Tree Size with Duplicate Detection
Estimating Search Tree Size with Duplicate Detection Levi H. S. Lelis Departamento de Informática Universidade Federal de Viçosa Viçosa, Brazil Roni Stern Information Systems Engineering Ben Gurion University
More informationEarly Work on Optimization-Based Heuristics for the Sliding Tile Puzzle
Planning, Search, and Optimization: Papers from the 2015 AAAI Workshop Early Work on Optimization-Based Heuristics for the Sliding Tile Puzzle Ariel Felner ISE Department Ben-Gurion University Israel felner@bgu.ac.il
More informationInformed Search Methods
Informed Search Methods How can we improve searching strategy by using intelligence? Map example: Heuristic: Expand those nodes closest in as the crow flies distance to goal 8-puzzle: Heuristic: Expand
More informationEnhanced Partial Expansion A*
Journal of Artificial Intelligence Research 50 (2014) 141-187 Submitted 9/13; published 5/14 Enhanced Partial Expansion A* Meir Goldenberg Ariel Felner Roni Stern Guni Sharon Ben-Gurion University of the
More informationPartial-expansion A* with Selective Node Generation
Partial-expansion A* with Selective Node Generation Ariel Felner Nathan Sturtevant Meir Goldenberg CS Department Guni Sharon University of Denver (USA) Roni Stern, Tal Beja Sturtevant@cs.du.edu ISE Department
More informationHeuristic Functions. Lecture Heuristic Search. Stefan Edelkamp
Heuristic Functions Lecture Heuristic Search Stefan Edelkamp 1 Overview Definition of heuristic estimator functions Admissible heuristics Consistent heuristics Heuristics in Puzzles: (n 2 1)-Puzzle and
More informationCE 473: Artificial Intelligence. Autumn 2011
CE 473: Artificial Intelligence Autumn 2011 A* Search Luke Zettlemoyer Based on slides from Dan Klein Multiple slides from Stuart Russell or Andrew Moore Today A* Search Heuristic Design Graph search Recap:
More informationBest-First Frontier Search with Delayed Duplicate Detection
Best-First Frontier Search with Delayed Duplicate Detection Richard E. Korf Computer Science Department University of California, Los Angeles Los Angeles, CA 90095 korf@cs.ucla.edu Abstract Best-first
More informationBasic Search Algorithms
Basic Search Algorithms Tsan-sheng Hsu tshsu@iis.sinica.edu.tw http://www.iis.sinica.edu.tw/~tshsu 1 Abstract The complexities of various search algorithms are considered in terms of time, space, and cost
More informationBootstrap Learning of Heuristic Functions
Proceedings of the Third Annual Symposium on Combinatorial Search (SOCS-10) Bootstrap Learning of Heuristic Functions Shahab Jabbari Arfaee Computing Science Department University of Alberta Edmonton,
More informationInformed Search. CS 486/686 University of Waterloo May 10. cs486/686 Lecture Slides 2005 (c) K. Larson and P. Poupart
Informed Search CS 486/686 University of Waterloo May 0 Outline Using knowledge Heuristics Best-first search Greedy best-first search A* search Other variations of A* Back to heuristics 2 Recall from last
More informationHeuristic Search. Robert Platt Northeastern University. Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA
Heuristic Search Robert Platt Northeastern University Some images and slides are used from: 1. CS188 UC Berkeley 2. RN, AIMA Recap: What is graph search? Start state Goal state Graph search: find a path
More informationITCS 6150 Intelligent Systems. Lecture 5 Informed Searches
ITCS 6150 Intelligent Systems Lecture 5 Informed Searches Informed Searches We are informed (in some way) about future states and future paths We use this information to make better decisions about which
More informationCMPUT 396 Sliding Tile Puzzle
CMPUT 396 Sliding Tile Puzzle Sliding Tile Puzzle 2x2 Sliding Tile States Exactly half of the states are solvable, the other half are not. In the case of 2x2 puzzles, I can solve it if I start with a configuration
More informationUsing Lookaheads with Optimal Best-First Search
Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence (AAAI-10) Using Lookaheads with Optimal Best-First Search Roni Stern Tamar Kulberis Ariel Felner Information Systems Engineering
More informationCommon Misconceptions Concerning Heuristic Search
Common Misconceptions Concerning Heuristic Search Robert C. Holte Computing Science Department, University of Alberta Edmonton, Canada T6G 2E8 (holte@cs.ualberta.ca Abstract This paper examines the following
More informationMove Pruning and Duplicate Detection
Move Pruning and Duplicate Detection Robert C. Holte Computing Science Department University of Alberta Edmonton, AB Canada T6G 2E8 rholte@ualberta.ca Abstract. This paper begins by showing that Burch
More informationThe Compression Power of Symbolic Pattern Databases
Proceedings of the Eighteenth International Conference on Automated Planning and Scheduling (ICAPS 2008) The Compression Power of Symbolic Pattern Databases Marcel Ball and Robert C. Holte Department of
More informationPuzzle Solving (single-agent search) Robert Holte Compu.ng Science Department University of Alberta
Puzzle Solving (single-agent search) Robert Holte Compu.ng Science Department University of Alberta 1 Puzzle = 1-player game Examples: Rubik s Cube Sliding-.le puzzle In this talk puzzles have determinis.c
More informationInformed search. Soleymani. CE417: Introduction to Artificial Intelligence Sharif University of Technology Spring 2016
Informed search CE417: Introduction to Artificial Intelligence Sharif University of Technology Spring 2016 Soleymani Artificial Intelligence: A Modern Approach, Chapter 3 Outline Best-first search Greedy
More informationVISUALIZING OBJECT AND METHOD GRANULARITY FOR PROGRAM PARALLELIZATION
VISUALIZING OBJECT AND METHOD GRANULARITY FOR PROGRAM PARALLELIZATION WILLIAM HUI, STEVE MACDONALD, JONATHAN SCHAEFFER AND DUANE SZAFRON Department of Computing Science, University of Alberta {stevem,duane,jonathan
More informationToday. Informed Search. Graph Search. Heuristics Greedy Search A* Search
Informed Search [These slides were created by Dan Klein and Pieter Abbeel for CS188 Intro to AI at UC Berkeley. All CS188 materials are available at http://ai.berkeley.edu.] Today Informed Search Heuristics
More informationUniversity of Alberta. Richard Valenzano. Master of Science. Department of Computing Science
University of Alberta SIMULTANEOUSLY SEARCHING WITH MULTIPLE ALGORITHM SETTINGS: AN ALTERNATIVE TO PARAMETER TUNING FOR SUBOPTIMAL SINGLE-AGENT SEARCH by Richard Valenzano A thesis submitted to the Faculty
More informationCS 343: Artificial Intelligence
CS 343: Artificial Intelligence Informed Search Prof. Scott Niekum University of Texas at Austin [These slides based on ones created by Dan Klein and Pieter Abbeel for CS188 Intro to AI at UC Berkeley.
More informationInformed search algorithms
Artificial Intelligence Topic 4 Informed search algorithms Best-first search Greedy search A search Admissible heuristics Memory-bounded search IDA SMA Reading: Russell and Norvig, Chapter 4, Sections
More informationLogistics. Heuristics & Constraint Satisfaction. Symbolic Integration. 573 Topics. Symbolic Integration. Search. E.g. x 2 e x dx = e x (x 2-2x+2) + C
Logistics Heuristics & Constraint Satisfaction CSE 573 University of Washington Reading for Monday Ch 6 (Game playing) Guest Speaker Henry Kautz Mini Projects A. Game Playing Choose your own game Experiment
More informationBlock-Parallel IDA* for GPUs
Proceedings of the Tenth International Symposium on Combinatorial Search (SoCS 2017) Block-Parallel IDA* for GPUs Satoru Horie, Alex Fukunaga Graduate School of Arts and Sciences The University of Tokyo
More informationInformed Search. Xiaojin Zhu Computer Sciences Department University of Wisconsin, Madison
Informed Search Xiaojin Zhu jerryzhu@cs.wisc.edu Computer Sciences Department University of Wisconsin, Madison [Based on slides from Andrew Moore http://www.cs.cmu.edu/~awm/tutorials ] slide 1 Main messages
More informationArtificial Intelligence Informed Search
Artificial Intelligence Informed Search Instructors: David Suter and Qince Li Course Delivered @ Harbin Institute of Technology [Many slides adapted from those created by Dan Klein and Pieter Abbeel for
More informationHeuris'c Search. Reading note: Chapter 4 covers heuristic search.
Heuris'c Search Reading note: Chapter 4 covers heuristic search. Credits: Slides in this deck are drawn from or inspired by a multitude of sources including: Shaul Markovitch Jurgen Strum Sheila McIlraith
More informationEfficient AND/OR Search Algorithms for Exact MAP Inference Task over Graphical Models
Efficient AND/OR Search Algorithms for Exact MAP Inference Task over Graphical Models Akihiro Kishimoto IBM Research, Ireland Joint work with Radu Marinescu and Adi Botea Outline 1 Background 2 RBFAOO
More informationInformed Search A* Algorithm
Informed Search A* Algorithm CE417: Introduction to Artificial Intelligence Sharif University of Technology Spring 2018 Soleymani Artificial Intelligence: A Modern Approach, Chapter 3 Most slides have
More informationClass Overview. Introduction to Artificial Intelligence COMP 3501 / COMP Lecture 2. Problem Solving Agents. Problem Solving Agents: Assumptions
Class Overview COMP 3501 / COMP 4704-4 Lecture 2 Prof. JGH 318 Problem Solving Agents Problem Solving Agents: Assumptions Requires a goal Assume world is: Requires actions Observable What actions? Discrete
More informationChapter 11 Search Algorithms for Discrete Optimization Problems
Chapter Search Algorithms for Discrete Optimization Problems (Selected slides) A. Grama, A. Gupta, G. Karypis, and V. Kumar To accompany the text Introduction to Parallel Computing, Addison Wesley, 2003.
More informationAnalyzing the Performance of Pattern Database Heuristics
Analyzing the Performance of Pattern Dataase Heuristics Richard E. Korf Computer Science Department University of California, Los Angeles Los Angeles, CA 90095 korf@cs.ucla.edu Astract We introduce a model
More informationHeuristic Search. Rob Platt Northeastern University. Some images and slides are used from: AIMA
Heuristic Search Rob Platt Northeastern University Some images and slides are used from: AIMA Recap: What is graph search? Start state Goal state Graph search: find a path from start to goal what are the
More informationArtificial Intelligence Informed search. Peter Antal
Artificial Intelligence Informed search Peter Antal antal@mit.bme.hu 1 Informed = use problem-specific knowledge Which search strategies? Best-first search and its variants Heuristic functions? How to
More informationInformed search methods
Informed search methods Tuomas Sandholm Computer Science Department Carnegie Mellon University Read Section 3.5-3.7 of Russell and Norvig Informed Search Methods Heuristic = to find, to discover Heuristic
More informationLecture 9. Heuristic search, continued. CS-424 Gregory Dudek
Lecture 9 Heuristic search, continued A* revisited Reminder: with A* we want to find the best-cost (C ) path to the goal first. To do this, all we have to do is make sure our cost estimates are less than
More informationInformed Search Algorithms
Informed Search Algorithms CITS3001 Algorithms, Agents and Artificial Intelligence Tim French School of Computer Science and Software Engineering The University of Western Australia 2017, Semester 2 Introduction
More informationSimultaneously Searching with Multiple Settings: An Alternative to Parameter Tuning for Suboptimal Single-Agent Search Algorithms
Simultaneously Searching with Multiple Settings: An Alternative to Parameter Tuning for Suboptimal Single-Agent Search Algorithms Richard Valenzano, Nathan Sturtevant, Jonathan Schaeffer University of
More informationDFS* and the Traveling Tournament Problem. David C. Uthus, Patricia J. Riddle, and Hans W. Guesgen
DFS* and the Traveling Tournament Problem David C. Uthus, Patricia J. Riddle, and Hans W. Guesgen Traveling Tournament Problem Sports scheduling combinatorial optimization problem. Objective is to create
More informationCS 343H: Artificial Intelligence
CS 343H: Artificial Intelligence Lecture 4: Informed Search 1/23/2014 Slides courtesy of Dan Klein at UC-Berkeley Unless otherwise noted Today Informed search Heuristics Greedy search A* search Graph search
More informationHeuristic Search INTRODUCTION TO ARTIFICIAL INTELLIGENCE BIL 472 SADI EVREN SEKER
Heuristic Search INTRODUCTION TO ARTIFICIAL INTELLIGENCE BIL 7 SADI EVREN SEKER Review of Search - BFS Review of Search - DFS Review of Search - IDDFS DFS Remembering: IDDFS(root, A, B, goal) D, F, E,
More informationAdvances in Parallel Branch and Bound
ECAI 2012 Advances in Parallel Branch and Bound Lars Otten and Rina Dechter Dept. of Computer Science University of California, Irvine Summary Parallelizing AND/OR Branch and Bound: - Advanced optimization
More informationA Pattern Database Approach for Solving the TopSpin Puzzle Problem
A Pattern Database Approach for Solving the TopSpin Puzzle Problem Pier Paolo Bortoluzzi University of Basel Examinar: Malte Helmert Supervisor: Martin Wehrle Abstract Finding optimal solutions for general
More informationChapters 3-5 Problem Solving using Search
CSEP 573 Chapters 3-5 Problem Solving using Search First, they do an on-line search CSE AI Faculty Example: The 8-puzzle Example: The 8-puzzle 1 2 3 8 4 7 6 5 1 2 3 4 5 6 7 8 2 Example: Route Planning
More informationAdvances In Industrial Logic Synthesis
Advances In Industrial Logic Synthesis Luca Amarù, Patrick Vuillod, Jiong Luo Design Group, Synopsys Inc., Sunnyvale, California, USA Design Group, Synopsys, Grenoble, FR Logic Synthesis Y
More informationCSCI 446: Artificial Intelligence
CSCI 446: Artificial Intelligence Informed Search Instructor: Michele Van Dyne [These slides were created by Dan Klein and Pieter Abbeel for CS188 Intro to AI at UC Berkeley. All CS188 materials are available
More informationTime complexity of iterative-deepening-a
Artificial Intelligence 129 (2001) 199 218 Time complexity of iterative-deepening-a Richard E. Korf a,, Michael Reid b, Stefan Edelkamp c a Computer Science Department, University of California, Los Angeles,
More informationData cube and OLAP. Selecting Views to Materialize. Aggregation view lattice. Selecting views to materialize. Limitations of static approach.
Data cube and OLAP Selecting Views to Materialize CPS 296.1 Topics in Database Systems Example data cube schema: Sale(store, product, customer, quantity) Store, product, customer are dimension attributes
More informationA Hybrid Recursive Multi-Way Number Partitioning Algorithm
Proceedings of the Twenty-Second International Joint Conference on Artificial Intelligence A Hybrid Recursive Multi-Way Number Partitioning Algorithm Richard E. Korf Computer Science Department University
More informationDeveloping MapReduce Programs
Cloud Computing Developing MapReduce Programs Dell Zhang Birkbeck, University of London 2017/18 MapReduce Algorithm Design MapReduce: Recap Programmers must specify two functions: map (k, v) * Takes
More informationHeuristic Search. Heuristic Search. Heuristic Search. CSE 3401: Intro to AI & LP Informed Search
CSE 3401: Intro to AI & LP Informed Search Heuristic Search. Required Readings: Chapter 3, Sections 5 and 6, and Chapter 4, Section 1. In uninformed search, we don t try to evaluate which of the nodes
More informationGeForce4. John Montrym Henry Moreton
GeForce4 John Montrym Henry Moreton 1 Architectural Drivers Programmability Parallelism Memory bandwidth 2 Recent History: GeForce 1&2 First integrated geometry engine & 4 pixels/clk Fixed-function transform,
More informationarxiv: v1 [cs.ai] 8 May 2017
Block-Parallel IDA* for GPUs (Extended Manuscript) Satoru Horie and Alex Fukunaga Graduate School of Arts and Sciences The University of Tokyo arxiv:1705.02843v1 [cs.ai] 8 May 2017 Abstract We investigate
More informationCoarse-to-Fine Search Techniques
Coarse-to-Fine Search Techniques by Ken Anderson, Nathan Sturtevant, Robert Holte, and Jonathan Schaeffer Technical Report TR 08-05 April 2008 DEPARTMENT OF COMPUTING SCIENCE University of Alberta Edmonton,
More informationCS 188: Artificial Intelligence. Recap Search I
CS 188: Artificial Intelligence Review of Search, CSPs, Games DISCLAIMER: It is insufficient to simply study these slides, they are merely meant as a quick refresher of the high-level ideas covered. You
More informationSearch Algorithms for Discrete Optimization Problems
Search Algorithms for Discrete Optimization Problems Ananth Grama, Anshul Gupta, George Karypis, and Vipin Kumar To accompany the text ``Introduction to Parallel Computing'', Addison Wesley, 2003. Topic
More informationIterative-Expansion A*
Iterative-Expansion A* Colin M. Potts and Kurt D. Krebsbach Department of Mathematics and Computer Science Lawrence University, Appleton, Wisconsin 54911 {colin.m.potts, kurt.krebsbach}@lawrence.edu Abstract
More information! A* is best first search with f(n) = g(n) + h(n) ! Three changes make it an anytime algorithm:
Anytime A* Today s s lecture Lecture 5: Search - 4 Hierarchical A* Jiaying Shen CMPSCI 683 Fall 2004 Other Examples of Hierarchical Problem Solving Reviews of A* and its varations 2 Anytime algorithms
More informationEECS 492 Midterm #1. Example Questions. Note: Not a complete exam!
EECS 492 Midterm #1 Example Questions Note: Not a complete exam! General Instructions This exam is closed book, except that you are allowed to refer to a single sheet of paper. You may use a calculator
More informationOptimal-Generation Variants of EPEA*
Optimal-Generation Variants of EPEA* Meir Goldenberg, Ariel Felner ISE Department Ben-Gurion University Israel mgoldenbe@gmail.com, felner@bgu.ac.il Nathan Sturtevant CS Department University of Denver
More informationCS 563 Advanced Topics in Computer Graphics QSplat. by Matt Maziarz
CS 563 Advanced Topics in Computer Graphics QSplat by Matt Maziarz Outline Previous work in area Background Overview In-depth look File structure Performance Future Point Rendering To save on setup and
More informationFringe Search: Beating A* at Pathfinding on Game Maps
Fringe Search: Beating A* at Pathfinding on Game Maps Yngvi Björnsson Markus Enzenberger, Robert C. Holte and Jonathan Schaeffer School of Computer Science Department of Computing Science Reykjavik University
More informationThe Spurious Path Problem in Abstraction
The Spurious Path Problem in Abstraction Gaojian Fan and Robert C. Holte University of Alberta, Edmonton, Alberta, Canada {gaojian, rholte}@ualberta.ca Abstract Abstraction is a powerful technique in search
More informationRobust Bidirectional Search via Heuristic Improvement
Robust via Heuristic Improvement Christopher Wilt and Wheeler Ruml Thanks to Michael Leighton, NSF (grants 0812141 and 1150068) and DARPA (grant N10AP20029) Christopher Wilt (UNH) Robust via Heuristic
More informationInformed (Heuristic) Search. Idea: be smart about what paths to try.
Informed (Heuristic) Search Idea: be smart about what paths to try. 1 Blind Search vs. Informed Search What s the difference? How do we formally specify this? A node is selected for expansion based on
More informationFront-to-End Bidirectional Heuristic Search with Near-Optimal Node Expansions
Front-to-End Bidirectional Heuristic Search with Near-Optimal Node Expansions Jingwei Chen University of Denver Denver, CO, USA jingchen@cs.du.edu Robert C. Holte University of Alberta Edmonton, AB, Canada
More informationCS 188: Artificial Intelligence
CS 188: Artificial Intelligence Today Informed Search Informed Search Heuristics Greedy Search A* Search Instructor: Marco Alvarez University of Rhode Island (These slides were created/modified by Dan
More informationParallel Programming. Parallel algorithms Combinatorial Search
Parallel Programming Parallel algorithms Combinatorial Search Some Combinatorial Search Methods Divide and conquer Backtrack search Branch and bound Game tree search (minimax, alpha-beta) 2010@FEUP Parallel
More informationBounded Suboptimal Search in Linear Space: New Results
Proceedings of the Seventh Annual Symposium on Combinatorial Search (SoCS 2014) Bounded Suboptimal Search in Linear Space: New Results Matthew Hatem and Wheeler Ruml Department of Computer Science University
More information