Many real life optimization problems contain both hard and soft constraints, as well as qualitative conditional preferences. However, there is no single formalism to specify all three kinds of information. We therefore propose a framework, based on both CP-nets and soft constraints, that handles both hard and soft constraints as well as conditional preferences efficiently and uniformly. We study the complexity of testing the consistency of preference statements, and show how soft constraints can faithfully approximate the semantics of conditional preference statements whilst improving the computational complexity. 1

In cooperative problem solving, the communication necessary for solution search can also lead to privacy loss on the part of the agents involved. Such loss can be assessed either by directly tallying the number and importance of specific items of information revealed or by tracking reductions in the set of possible values associated with a particular item of information. In both cases information loss can occur either because of direct communication or by inferences that other agents make from one’s communications. The results of these inferences are stored in the “views ” that agents have of other agents. In the present work on distributed constraint solving, such views are organized as extensions of normal CSP representations that model information about possible values in unknown CSPs of other agents. Here we show how this approach can be extended so that agents also maintain views of other agents ’ views of themselves; the latter are called “mirror views”. Mirror views can in turn be used to monitor one’s own privacy loss, and can support strategies designed to reduce the loss of particular kinds of private information. Experiments with a simulated meeting scheduling system show that it is possible to reduce privacy loss with strategies based on mirror views.

Security protocols stipulate how the remote principals of a computer network should interact in order to obtain specific security goals. The crucial goals of confidentiality and authentication may be achieved in various forms, each of different strength. Using soft (rather than crisp) constraints, we develop a uniform formal notion for the two goals. They are no longer formalised as mere yes/no properties as in the existing literature, but gain an extra parameter, the security level. For example, different messages can enjoy different levels of confidentiality, or a principal can achieve different levels of authentication with different principals. The goals are formalised within a general framework for protocol analysis that is amenable to mechanisation by model checking. Following the application of the framework to analysing the asymmetric Needham-Schroeder protocol (Bella and Bistarelli 2001; Bella and Bistarelli 2002), we have recently discovered a new attack on that protocol as a form of retaliation by principals who have been attacked previously. Having commented on that attack, we then demonstrate the framework on a bigger, largely deployed protocol consisting of three phases, Kerberos.

Soft constraints are very exible and expressive. However, they also are very complex to handle. For this reason, it may be reasonable in several cases to pass to an abstract version of a given soft constraint problem, and then to bring some useful information from the abstract problem to the concrete one. This will hopefully make the search for a solution, or for an optimal solution, of the concrete problem, faster.

We propose an abstraction scheme for soft constraint problems and we study its main properties. Processing the abstracted version of a soft constraint problem can help us in many ways: for example, to nd good approximations of the optimal solutions, or also to provide us with information that can make the subsequent search for the best solution easier. The results of this paper show that the proposed scheme is promising; thus they can be used as a stable formal base for any experimental work specific to a particular class of soft constraint problems.

Abstract not found

Real-life dynamic problems may lead to inconsistent constraints systems for which a solution must be found even if constraints have to be relaxed. In this paper, we propose a best-first search to handle such problems. Classical backtracking search algorithms are extended in twoways: identification of good backtrackpoints as in Intelligent Backtracking techniques and maximum use of independant work (that would have been discarded with a mere backtrack). We first describe an operational semantics for our search method. Then we specialize it to handle constraint relaxation over finite domains. The practical use of this approach is demonstrated by theoretical complexity analysis and experiments.

. Soft constraints are very flexible and expressive. However, they also are very complex to handle. For this reason, it may reasonable in several cases to pass to an abstract version of a given soft problem, and then to bring some useful information from the abstract problem to the concrete one. This will hopefully make the search for a solution, or for an optimal solution, of the concrete problem, faster. In this paper we review the main concepts and properties of our abstraction framework for soft constraints, and we show how it can be used to import constraint propagation algorithms from the abstract scenario to the concrete one. This may be useful when we don't have any (or any e#cient) propagation algorithm in the concrete setting. 1

Abstract. Many real life optimization problems are defined in terms of both hard and soft constraints, and qualitative conditional preferences. However, there is as yet no single framework for combined reasoning about these three kinds of information. In this paper we study how to exploit classical and soft constraint solvers for handling qualitative preference statements such as those captured by the CP-nets model. In particular, we show how hard constraints are sufficient to model the optimal outcomes of a possibly cyclic CP-net, and how soft constraints can faithfully approximate the semantics of acyclic conditional preference statements whilst improving the computational efficiency of reasoning about these statements. 1

Constraint Violation Minimization Problems arise when dealing with over-constrained CSPs. Unfortunately, experiments and practice show that they quickly become too large and too difficult to be optimally solved. In this context, multiple methods (limited tree search, heuristic or stochastic local search) are available to produce non-optimal, but good quality solutions, and thus to provide the user with anytime upper bounds of the problem optimum. On the other hand, few methods are available to produce anytime lower bounds of this optimum.