Learning While Searching in Constraint-Satisfaction-Problems
نویسنده
چکیده
The popular use of backtracking as a control strategy for theorem proving in PROLOG and in Truth-MaintenanceSystems (TMS) led to increased interest in various schemes for enhancing the efficiency of backtrack search. Researchers have referred to these enhancement schemes by the names ‘ ‘Intelligent Backtracking’ ’ (in PROLOG), ‘ ‘Dependencydirected-backtracking” (in TMS) and others. Those improvements center on the issue of “jumping-back” to the source of the problem in front of dead-end situations. This paper examines another issue (much less explored) which arises in dead-ends. Specifically, we concentrate on the idea of constraint recording, namely, analyzing and storing the reasons for the dead-ends, and using them to guide future decisions, so that the same conflicts will not arise again. We view constraint recording as a process of learning, and examine several possible learning schemes studying the tradeoffs between the amount of learning and the improvement in search efficiency.
منابع مشابه
Ng-backmarking - an algorithm for constraint satisfaction
Ng-backmarking with Min-conflict repair, a hybrid algorithm for solving constraint satisfaction problems, is presented in the context of the four main approaches to constraint satisfaction and optimisation: tree-search, domainfiltering, solution repair, and learning while searching. Repair-based techniques are often designed to use local gradients to direct the search for a solution to a constr...
متن کاملAutomatic Generation of Implied Constraints
A well-known difficulty with solving Constraint Satisfaction Problems (CSPs) is that, while one formulation of a CSP may enable a solver to solve it quickly, a different formulation may take prohibitively long to solve. We demonstrate a system for automatically reformulating CSP solver models by combining the capabilities of machine learning and automated theorem proving with CSP systems. Our s...
متن کاملLearning Constraint Satisfaction Problems: An ILP Perspective
We investigate the problem of learning constraint satisfaction problems from an inductive logic programming perspective. Constraint satisfaction problems are the underlying basis for constraint programming and there is a long standing interest in techniques for learning these. Constraint satisfaction problems are often described using a relational logic, so inductive logic programming is a natu...
متن کاملSmall is Beautiful: A Brute-Force Approach to Learning First-Order Formulas
We describe a method for learning formulas in firstorder logic using a brute-force, smallest-first search. The method is exceedingly simple. It generates all irreducible well-formed formulas up to a fixed size and tests them against a set of examples. Although the method has some obvious limitations due to its computational complexity, it performs surprisingly well on some tasks. This paper des...
متن کاملAutomated within-problem learning for Constraint Satisfaction Problems
Within-problem learning, and in particular learning from failure, has proven to be extremely beneficial in solving combinatorial problems in the boolean satisfiability domain (SAT) where both clause learning and clause weighting are used by many of the top SAT solvers. Similar research in solving constraint satisfaction problems (CSPs) has been done on nogood learning and constraint weighting (...
متن کاملWhere the Exceptionally Hard Problems Are 1
Constraint satisfaction problems exhibit a phase transition as a problem parameter is varied, from a region where most problems are easy and soluble to a region where most problems are easy but insoluble. In the intervening phase transition region, the median problem diiculty is greatest. However, in the easy and soluble region, occasional exceptionally hard problems (ehps) can be found, which ...
متن کامل