Filters








12,004 Hits in 2.0 sec

The Recomputation Manifesto [article]

Ian P. Gent
2013 arXiv   pre-print
I also thank more recent colleagues, including Edwin Brady, Chris Jefferson, Steve Linton, Ian Miguel, Pete Nightingale, Karen Petrie, Aaron Quigley, and Jonathan Ward.  ...  Ian P. Gent School of Computer Science, University of St Andrews, Fife, KY16 9SX, Scotland, UK. ian.gent@st-andrews.ac.uk. http://www.cs.st-andrews.ac.uk/~ipg  ... 
arXiv:1304.3674v1 fatcat:evrgfqfy3zhjbegn2v774mc5su

Reliability of Computational Experiments on Virtualised Hardware [article]

Ian P. Gent, Lars Kotthoff
2011 arXiv   pre-print
-A Balanced Incomplete Block Design (BIBD) problem that takes about a minute to solve, CSPLib (Gent and Walsh, 1999) problem 028.  ...  Experimental evaluation To evaluate the reliability of experimental results, we used the Minion constraint solver (Gent et al., 2006) . We ran it on the following three problems.  ... 
arXiv:1110.6288v1 fatcat:kubk2bi5uvcfll5ea2n4g7tfba

Towards Reformulating Essence Specifications for Robustness [article]

Özgür Akgün, Alan M. Frisch, Ian P. Gent, Christopher Jefferson, Ian Miguel, Peter Nightingale, András Z. Salamon
2021 arXiv   pre-print
The Essence language allows a user to specify a constraint problem at a level of abstraction above that at which constraint modelling decisions are made. Essence specifications are refined into constraint models using the Conjure automated modelling tool, which employs a suite of refinement rules. However, Essence is a rich language in which there are many equivalent ways to specify a given problem. A user may therefore omit the use of domain attributes or abstract types, resulting in fewer
more » ... ulting in fewer refinement rules being applicable and therefore a reduced set of output models from which to select. This paper addresses the problem of recovering this information automatically to increase the robustness of the quality of the output constraint models in the face of variation in the input Essence specification. We present reformulation rules that can change the type of a decision variable or add attributes that shrink its domain. We demonstrate the efficacy of this approach in terms of the quantity and quality of models Conjure can produce from the transformed specification compared with the original.
arXiv:2111.00821v1 fatcat:mu3efptrozdcdaakkerxw24ua4

Modelling Constraint Solver Architecture Design as a Constraint Problem [article]

Ian P. Gent and Chris Jefferson and Lars Kotthoff and Ian Miguel
2011 arXiv   pre-print
eq ( p v w p r o v i d e s [ 1 ] , 0 ) r e i f y ( watched−o r ({ eq ( pvw , 1 ) , eq ( pvw , 2 ) , eq ( pvw , 3 ) } ) , p v w p r o v i d e s [ 4 ] ) eq ( p v w p r o v i d e s [ 5 ] , 0 ) eq ( p v w  ...  p r o v i d e s [ 2 ] , 0 ) r e i f y ( watched−o r ({ eq ( pvw , 3 ) } ) , p v w p r o p e r t i e s [ 2 ] ) r e i f y ( watched−o r ({ eq ( pvw , 2 ) } ) , p v w p r o p e r t i e s [ 1 ] ) r e i f  ... 
arXiv:1110.6290v1 fatcat:fhkic2c4brbgpbvtwghaq3hzvm

Lazy Explanations for Constraint Propagators [chapter]

Ian P. Gent, Ian Miguel, Neil C. A. Moore
2010 Lecture Notes in Computer Science  
Explanations are a technique for reasoning about constraint propagation, which have been applied in many learning, backjumping and user-interaction algorithms for constraint programming. To date explanations for constraints have usually been recorded eagerly when constraint propagation happens, which leads to inecient use of time and space, because many will never be used. In this paper we show that it is possible and highly eective to calculate explanations retrospectively when they are
more » ... hen they are needed. To this end, we implement lazy explanations in a state of the art learning framework. Experimental results conrm the eectiveness of the technique: we achieve reduction in the number of explanations calculated up to a factor of 200 and reductions in overall solve time up to a factor of 5.
doi:10.1007/978-3-642-11503-5_19 fatcat:h32w7yc33nhlxheq3ssvncqaze

Generating custom propagators for arbitrary constraints

Ian P. Gent, Christopher Jefferson, Steve Linton, Ian Miguel, Peter Nightingale
2014 Artificial Intelligence  
Gent and Smith also proposed a variable and value ordering that we use here. s[n] ∈ {−1, 1}.  ...  Moves to state 3 State 3 (Stored State A) Apply(i) = P 3 [i] Update(q) : Calls compose(P 3 , q, P 4 ).  ... 
doi:10.1016/j.artint.2014.03.001 fatcat:pufj7ybntnenhb3fs5pjei6lna

Conditional Symmetry Breaking [chapter]

Ian P. Gent, Tom Kelsey, Steve A. Linton, Iain McDonald, Ian Miguel, Barbara M. Smith
2005 Lecture Notes in Computer Science  
Ian Gent is supported by a Royal Society of Edinburgh SEELLD/RSE Support Research Fellowship. Ian Miguel is supported by a UK Royal Academy of Engineering/EPSRC Research Fellowship.  ...  A conditional symmetry of a CSP P holds only in a sub-problem P of P . The conditions of the symmetry are the constraints necessary to generate P from P .  ... 
doi:10.1007/11564751_21 fatcat:lvxwh2aaqfgkjdfwr3cicdgjp4

Metamorphic Testing of Constraint Solvers [chapter]

Özgür Akgün, Ian P. Gent, Christopher Jefferson, Ian Miguel, Peter Nightingale
2018 Lecture Notes in Computer Science  
Then the result S Q of applying q i to S must be contained in M as q i is inflationary, and must satisfy N P ⊆ V S Q , because N P ⊆ S, and N P is a fixed point for P and therefore all of the p i .  ...  If N P is a result of applying elements of P to N until a fixed point is reached, and similarly for M Q , then N P ⊆ V M Q .  ... 
doi:10.1007/978-3-319-98334-9_46 fatcat:3isau54rvzewdhyvtyhvaarxre

Watched Literals for Constraint Propagation in Minion [chapter]

Ian P. Gent, Chris Jefferson, Ian Miguel
2006 Lecture Notes in Computer Science  
Ian Miguel is supported by a UK Royal Academy of Engineering/EPSRC Fellowship.  ...  Where i ranges from 1 to n, the constraints are: V [P [2 * i]] = elem i V [P [2 * i + 1]] = elem i P [2 * i] = i + P [2 * i + 1] We found all solutions to Langford's problem up to n = 8 using this model  ...  For each i ∈ {1, 2, . . .} the 2i th and 2i + 1 st variables in P are the first and second positions of i in V . Each variable in P has domain {0, 1, . . . , 2n − 1}, indexing matrices from 0.  ... 
doi:10.1007/11889205_15 fatcat:wa5p54zhrfgczjy66za52rpgfy

An empirical study of learning and forgetting constraints

Ian P. Gent, Ian Miguel, Neil C.A. Moore
2012 AI Communications  
%, since the best P % constraints must do at least P % of propagations.  ...  In Table 1 for each chosen percentage P , we give what percentage of the best constraints are needed to account for P % of overall non-branching propagation 3 .  ... 
doi:10.3233/aic-2012-0524 fatcat:pmnlmyuz7jesrml3cl2lpg7jgu

The TSP phase transition

Ian P. Gent, Toby Walsh
1996 Artificial Intelligence  
Eq. (1) therefore provides a poor approximation for the mean optimal tour length for small n. 1 .P.  ...  Gent. 7: Walsh/Arti$cial Intelligence RR (1996) [349] [350] [351] [352] [353] [354] [355] [356] [357] [358] In addition, it provides no indication of how tour lengths are distributed around the mean  ... 
doi:10.1016/s0004-3702(96)00030-6 fatcat:26tkf6e4a5bunnbgs2gu46ynv4

The satisfiability constraint gap

Ian P. Gent, Toby Walsh
1996 Artificial Intelligence  
Dubois, P Andre, Y. Boutkhad and J. Carher, SAT versus UNSAT, Presented at the Second DIMACS Challenge Workshop (1993). 18 1 1.P Gent and T.  ...  Walsh, Easy problems are sometimes hard, ArtijI Intell. 70 (1994) 335-34.5. 191 1.P Gent and T. Walsh, The SAT phase transition, in: A.G.  ... 
doi:10.1016/0004-3702(95)00047-x fatcat:vpekfo6xkvgxjbu6l2nnuvr2gq

Qualitative modelling via constraint programming

Thomas W. Kelsey, Lars Kotthoff, Christopher A. Jefferson, Stephen A. Linton, Ian Miguel, Peter Nightingale, Ian P. Gent
2014 Constraints  
These definitions allow us to post qualitative constraints about peak populations ∃p ∈ [1, . . . , n] such that ∀i > p, X ′ [i] < 0 ∧ ∀i < p, X ′ [i] > 0.  ...  Others are not (depending on Lipschitz conditions and whether or not P = P SP ACE [32] ). Nonlinear ODEs are strictly harder to solve as a class, and most PDEs have no closed form solution.  ... 
doi:10.1007/s10601-014-9158-6 fatcat:5j3qjsfdmrdubb4acuaalrqki4

Qualitative Modelling via Constraint Programming: Past, Present and Future [article]

Thomas W. Kelsey, Lars Kotthoff, Christoffer A. Jefferson, Stephen A. Linton, Ian Miguel, Peter Nightingale, Ian P. Gent
2012 arXiv   pre-print
These definitions allow us to post qualitative constraints about peak populations ∃p ∈ [1, . . . , n] such that ∀i > p, X [i] < 0 ∧ ∀i < p, X [i] > 0.  ...  Others are not (depending on Lipschitz conditions and whether or not P = P SP ACE [29] ). Nonlinear ODEs are strictly harder to solve as a class, and most PDEs have no closed form solution.  ... 
arXiv:1209.3916v1 fatcat:s7w36cffofesdiga4zaxfwqt4y

Generating Special-Purpose Stateless Propagators for Arbitrary Constraints [chapter]

Ian P. Gent, Chris Jefferson, Ian Miguel, Peter Nightingale
2010 Lecture Notes in Computer Science  
Gent and Smith identified 7 symmetric images of the sequence [13] . We use these to post 7 symmetry-breaking constraints on s.  ...  Case Study: Low Autocorrelation Binary Sequences The Low Autocorrelation Binary Sequence (LABS) problem is described by Gent and Smith [13] .  ... 
doi:10.1007/978-3-642-15396-9_19 fatcat:pzv6uhazsfb7vhh5eyk7xf5s4i
« Previous Showing results 1 — 15 out of 12,004 results