#### DMCA

## GSAT and Dynamic Backtracking (1994)

### Cached

### Download Links

- [www.ai.mit.edu]
- [ftp.cs.unh.edu]
- [cs.ru.ac.za]
- [ftp.cirl.uoregon.edu]
- [ftp.cirl.uoregon.edu]
- [www.cs.cmu.edu]
- [www.stanford.edu]
- [www.jair.org]
- [jair.org]
- [arxiv.org]
- DBLP

### Other Repositories/Bibliography

Venue: | Journal of Artificial Intelligence Research |

Citations: | 384 - 15 self |

### Citations

5030 | Optimization by simulated annealing - Kirkpatrick, Gelatt - 1982 |

1249 | A computing procedure for quantification theory - Davis, Putnam - 1960 |

938 | Foundations of Constraint Satisfaction - Tsang - 1993 |

838 | Heuristics ±Intelligent Search Strategies for Computer Problem Solving - Pearl - 1984 |

825 | A machine program for theorem proving - Davis, Logemann, et al. - 1962 |

724 | A new method for solving hard satisfiability problems - Selman, Levesque, et al. - 1992 |

681 | Towards a theory of declarative knowledge - Apt, Blair, et al. - 1988 |

681 | A machine program for theorem-proving - Davis, Logemann, et al. - 1962 |

672 | Where the really hard problems are - Cheeseman, Kanefsky, et al. - 1991 |

572 | Pushing the envelope: Planning, propositional logic, and stochastic search - Kautz, Selman - 1996 |

492 | Semantical considerations on non-monotonic logic - Moore - 1985 |

330 | de Kleer. Foundations of assumption-based truth maintenance systems: Preliminary report - Reiter, Johan - 1987 |

312 | and Easy Distributions of SAT problems - Hard - 1992 |

311 | Local search strategies for satisfiability testing - Selman, Kautz, et al. - 1993 |

302 | Limited Discrepancy Search - Harvey, Ginsberg - 1995 |

295 | Practical Planning: Extending the Classical AI Planning Paradigm - Wilkins - 1988 |

284 |
Forward reasoning and dependencydirected backtracking in a system for computer-aided circuit analysis
- Stallman, Sussman
- 1977
(Show Context)
Citation Context ...hat have been proposed to this involvesnding ways to backtrack directly to some state that might actually allow us to make progress, in this case Arizona or earlier. Dependency-directed backtracking (=-=Stallman & Sussman, 1977-=-) involves a direct backtrack to the source of the diculty; backjumping (Gaschnig, 1979) avoids the computational overhead of this technique by using syntactic methods to estimate the point to which ... |

253 | Reasoning about change: time and causation from the stand point of artificial intelligent - Shoham - 1986 |

242 | The alternating fixpoint of logic programs with negation - Gelder - 1989 |

225 | Domain-independent extensions to GSAT: Solving large structured satisfiability problems - Selman, Kautz - 1993 |

225 | Semantical considerations on modal logic - Kripke - 1963 |

217 | Many hard examples for resolution - Chvátal, Szemerédi - 1988 |

217 | Iterative deepening: An optimal admissible tree search - Korf - 1985 |

208 | Experimental results on the cross-over point in satis problems - Crawford, Auton - 1993 |

194 | Performance measurement and analysis of certain search algorithms
- GASCHNIG
- 1979
(Show Context)
Citation Context ...ctually allow us to make progress, in this case Arizona or earlier. Dependency-directed backtracking (Stallman & Sussman, 1977) involves a direct backtrack to the source of the diculty; backjumping (=-=Gaschnig, 1979-=-) avoids the computational overhead of this technique by using syntactic methods to estimate the point to which backtrack is necessary. cs1993 AI Access Foundation and Morgan Kaufmann Publishers. All ... |

159 |
Solving large-scale constraintsatisfaction and scheduling problems using a heuristic repair method
- Minton, Philips, et al.
- 1990
(Show Context)
Citation Context ... Another way to look at this idea is that we have found a way to \erase" the value given to a variable directly as opposed to backtracking to it. This idea has also been explored by Minton et.al. in (=-=Minton, Johnston, Philips, & Laird, 1990-=-) and by Selman et.al. in (Selman, Levesque, & Mitchell, 1992); these authors also directly replace values assigned to variables in satisability problems. Unfortunately, the heuristic repair method u... |

142 | Experimental results on the crossover point in random 3-SAT - Crawford, Auton - 1996 |

139 | Negation as failure using tight derivations for general logic programs - Gelder - 1988 |

132 | Look-Ahead Techniques for Micro Opportunistic Job Shop Scheduling - Sadeh - 1991 |

128 | Experimental results on the application of satisfiability algorithms to scheduling problems - Crawford, Baker - 1994 |

120 | On stratified autoepistemic theories - Gelfond - 1987 |

120 | Explaining and repairing plans that fail - Hammond - 1990 |

106 | Slack-based Heuristics for Constraint Satisfaction Problems - Smith, Cheng - 1993 |

81 | A rearrangement search strategy for determining propositional satisfiability - Zabih, McAllester - 1988 |

77 | A guide to the modal logics of know-ledge and belief: a preliminary draft - Halpern, Moses - 1985 |

75 | Systematic and nonsystematic search strategies - Langley - 1992 |

74 | A complexity analysis of space-bound learning algorithms for the constraint satisfaction problem - Bayardo, Miranker - 1996 |

67 |
A new method for solving hard satis problems
- Selman, Levesque, et al.
- 1992
(Show Context)
Citation Context ...rase" the value given to a variable directly as opposed to backtracking to it. This idea has also been explored by Minton et.al. in (Minton, Johnston, Philips, & Laird, 1990) and by Selman et.al. in (=-=Selman, Levesque, & Mitchell, 1992-=-); these authors also directly replace values assigned to variables in satisability problems. Unfortunately, the heuristic repair method used is incomplete because no dependency information is retain... |

66 | Iterative broadening
- Ginsberg, Harvey
- 1992
(Show Context)
Citation Context ...ed between solution attempts and a maximum of 1000 backtracks were permitted before the program was deemed to have failed. In both cases, the algorithms were extended to include iterative broadening (=-=Ginsberg & Harvey, 1992-=-), the cheapest-rst heuristic and forward checking. Cheapest-rst has also been called \most constrainedsrst" and selects for instantiation that variable with the fewest number of remaining possibili... |

57 | Nonsystematic backtracking search - Harvey, Ginsberg - 1995 |

56 | Job shop scheduling by local search - Vaessens, Aarts, et al. - 1996 |

53 | Experimental evaluation of preprocessing techniques in constraint satisfaction problems
- Dechter, Meiri
- 1989
(Show Context)
Citation Context ...ned a value early in the search. 37 Ginsberg This reordering should not be confused with the work of authors who have suggested a dynamic choice among the variables that remain to be assigned values (=-=Dechter & Meiri, 1989-=-; Ginsberg et al., 1990; P. Purdom & Robertson, 1981; Zabih & McAllester, 1988); we are instead reordering the variables that have been assigned values in the search thus far. Another way to look at t... |

50 | Using deep structure to locate hard problems - Williams, Hogg - 1992 |

49 | Search lessons learned from crossword puzzles
- Ginsberg, Frank, et al.
- 1990
(Show Context)
Citation Context ...he problem in question does not split into two or more disjoint subproblems. 1 6. Experimentation Dynamic backtracking has been incorporated into the crossword-puzzle generation program described in (=-=Ginsberg, Frank, Halpin, & Torrance, 1990-=-), and leads to signicant performance improvements in that restricted domain. More specically, the method was tested on the problem of generating 19 puzzles of sizes ranging from 2 2 to 13 13; eac... |

41 | Generating feasible schedules under complex metric constraints - Cheng, Smith - 1994 |

41 |
A new method for solving constraint satisfaction problems
- Seidel
- 1981
(Show Context)
Citation Context ...e of 34 Dynamic Backtracking the automatic realization that the problem splits into disjoint subproblems. Other authors have also discussed the idea of applying divide-and-conquer techniques to csps (=-=Seidel, 1981-=-; Zabih, 1990), but their methods suer from the disadvantage that they constrain the order in which unassigned variables are assigned values, perhaps at odds with the common heuristic of assigning va... |

40 | Intelligent Backtracking Techniques for Job Shop Scheduling - Xiong, Sadeh, et al. - 1521 |

37 |
Ordering Conjunctive Queries
- Smith, Genesereth
- 1985
(Show Context)
Citation Context ...ed \most constrainedsrst" and selects for instantiation that variable with the fewest number of remaining possibilities (i.e., that variable for which it is cheapest to enumerate the possible values (=-=Smith & Genesereth, 1985-=-)). Forward checking prunes the set of possibilities for crossing words whenever a new word is entered and constitutes our experimental choice of eliminationmechanism: at any point, words for which th... |

34 |
Solving combinatorial search problems by intelligent backtracking
- Bruynooghe
- 1981
(Show Context)
Citation Context ...the problem. Backjumping avoids this problem by resetting the set E i of eliminating explanations in step 2 of Algorithm 3.3. The description that we have given is quite similar to that developed in (=-=Bruynooghe, 1981-=-). The explanations there are somewhat coarser than ours, listing all of the variables that have been involved in any eliminating explanation for a particular variable in the csp, but the idea is esse... |

27 | The application of theorem proving to question-answering systems - Green - 1969 |

18 | Easy to be hard: difficult problems for greedy algorithms - Konolige - 1994 |

17 | Backtracking with multi-level dynamic search rearrangement - Purdom, Brown, et al. - 1981 |

14 |
Some applications of graph bandwidth to constraint satisfaction problems
- Zabih
- 1990
(Show Context)
Citation Context ...c Backtracking the automatic realization that the problem splits into disjoint subproblems. Other authors have also discussed the idea of applying divide-and-conquer techniques to csps (Seidel, 1981; =-=Zabih, 1990-=-), but their methods suer from the disadvantage that they constrain the order in which unassigned variables are assigned values, perhaps at odds with the common heuristic of assigning valuessrst to t... |

13 | The MVL theorem proving system - Ginsberg - 1991 |

13 | Experimenting with new systematic and nonsystematic search techniques
- Jonsson, Ginsberg
- 1993
(Show Context)
Citation Context ...wo methods. 2 Although initially comparable, dynamic backtracking provides increasing computational savings as the problems become more dicult. A somewhat broader set of experiments is described in (=-=Jonsson & Ginsberg, 1993-=-) and leads to similar conclusions. There are some examples in (Jonsson & Ginsberg, 1993) where dynamic backtracking leads to performance degradation, however; a typical case appears in Figure 4. 3 In... |

13 | Logic programs with external procedures - nski, Bonnier, et al. |

11 | Hybrid Reasoning Using Universal Attachment - Myers - 1994 |

7 | User's guide to the MVL system - Ginsberg - 1989 |

6 | Constraintbased scheduling for batch manufacturing - Goldman, Boddy - 1997 |

6 | Partial order backtracking
- McAllester
- 1993
(Show Context)
Citation Context ...ne has already been eliminated for other reasons, backtracking to i is guaranteed to generate another backtrack immediately and should probably be avoided if possible. 4. Another solution appears in (=-=McAllester, 1993-=-). 39 Ginsberg Finally, there is some measure of the \directness" with which a variable bears on a problem. If we are unable tosnd a value for a particular variable i, it is probably sensible to backt... |

5 | Search and job shop scheduling - Harvey - 1994 |

2 | An algorithm for scheduling improvement by scheduling shifting - Fox - 1996 |

1 | Partial order backtracking. ftp.ai.mit.edu:/pub/dam/dynamic.ps - McAllester - 1993 |