## Preference Handling –- An Introductory Tutorial

Citations: | 15 - 0 self |

### BibTeX

@MISC{Brafman_preferencehandling,

author = {Ronen I. Brafman and Carmel Domshlak},

title = {Preference Handling –- An Introductory Tutorial },

year = {}

}

### OpenURL

### Abstract

We present a tutorial introduction to the area of preference handling – one of the core issues in the design of any system that automates or supports decision making. The main goal of this tutorial is to provide a framework, or perspective, within which current work on preference handling – representation, reasoning, and elicitation – can be understood. Our intention is not to provide a technical description of the diverse methods used, but rather, to provide a general perspective on the problem and its varied solutions and to highlight central ideas and techniques.

### Citations

7048 |
Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference
- Pearl
- 1988
(Show Context)
Citation Context ...ptimal assignment to a GAI value function/cost network/weighted CSP is identical to that of finding most probable explanation (MPE) in graphical probabilistic models such as Bayes Nets or Join-trees (=-=Pearl, 1988-=-; Lauritzen & Spiegelhalter, ). Indeed, there is general close correspondence between probabilistic reasoning and preferences. Techni10cally, a joint probability distribution over random variables x1... |

1423 |
A Logic for Default Reasoning
- Reiter
- 1980
(Show Context)
Citation Context ...he best known interpretation for non-monotonic logics is known as the preference semantics (Shoham, 1987), and thus most nonmonotonic logics, such as (Kraus, Lehmann, & Magidor, 1990; McCarthy, 1980; =-=Reiter, 1980-=-), can be viewed as expressing preference statements, usually associated with aggressive interpretations that attempt to deduce as much information as possible. Graphical models play an important role... |

1065 |
Introduction to Lattices and Order
- Davey, Priestley
- 2002
(Show Context)
Citation Context ...ersky, 1971; von Wright, 1963; Halldén, 1957), in economics and game-theory (K.J.Arrow & Raynaud, 1986; Debreu, 1954; von Neumann & Morgenstern, 1947; Fishburn, 1969), in mathematics (Birkhoff, 1948; =-=Davey & Priestley, 2002-=-), in operations research and decision analysis (Fishburn, 1974; Wald, 1950; Bouyssou, Marchant, Pirlot, Tsoukias, & Vincke, 2006), in psychology (Tversky, 1967, 1969; Kahneman & Tversky, 1979, 1984),... |

815 |
Circumscription: A Form of Nonmonotonic Reasoning
- McCarthy
- 1980
(Show Context)
Citation Context ...994). In fact, the best known interpretation for non-monotonic logics is known as the preference semantics (Shoham, 1987), and thus most nonmonotonic logics, such as (Kraus, Lehmann, & Magidor, 1990; =-=McCarthy, 1980-=-; Reiter, 1980), can be viewed as expressing preference statements, usually associated with aggressive interpretations that attempt to deduce as much information as possible. Graphical models play an ... |

560 |
Constraint Processing
- Dechter
- 2003
(Show Context)
Citation Context ...n the left-hand side we see a GAI value function, and on the right hand side, we see the corresponding GAI-network. In the constraint optimization literature, this structure is called a cost network (=-=Dechter, 2003-=-), and the problem of finding an assignment with maximal cost for such a network is well studied. Importantly, the algorithms used to solve this problem utilize the graphical structure of the network.... |

540 | Nonmonotonic reasoning, preferential models and cumulative logics
- Kraus, Lehmann, et al.
- 1990
(Show Context)
Citation Context ...yle & Wellman, 1994; Boutilier, 1994). In fact, the best known interpretation for non-monotonic logics is known as the preference semantics (Shoham, 1987), and thus most nonmonotonic logics, such as (=-=Kraus, Lehmann, & Magidor, 1990-=-; McCarthy, 1980; Reiter, 1980), can be viewed as expressing preference statements, usually associated with aggressive interpretations that attempt to deduce as much information as possible. Graphical... |

468 |
The Adaptive Decision Maker
- Payne, Bettman, et al.
- 1993
(Show Context)
Citation Context ... a focus on automating the process of preference elicitation (Ha & Haddawy, 1997, 1999; Torrens, Faltings, & Pu, 2002; Pu & Faltings, 2004; Faltings, Torrens, & Pu, 2004; Braziunas & Boutilier, 2005; =-=Payne, Bettman, & Johnson, 1993-=-; Smith & McGinty, 2003). Casting preference elicitation for policy optimization as a properly defined decision process was first suggested in (Chajewska, Koller, & Parr, 2000; Chajewska, Getoor, Norm... |

379 |
Utility Theory for Decision Making
- Fishburn
- 1969
(Show Context)
Citation Context ...ature (Hansson, 2001a; Krantz, Luce, Suppes, & Tversky, 1971; von Wright, 1963; Halldén, 1957), in economics and game-theory (K.J.Arrow & Raynaud, 1986; Debreu, 1954; von Neumann & Morgenstern, 1947; =-=Fishburn, 1969-=-), in mathematics (Birkhoff, 1948; Davey & Priestley, 2002), in operations research and decision analysis (Fishburn, 1974; Wald, 1950; Bouyssou, Marchant, Pirlot, Tsoukias, & Vincke, 2006), in psychol... |

375 | An Introduction to Kernel-Based Learning Algorithms
- Muller, Mika, et al.
- 2001
(Show Context)
Citation Context ...ssumptions of the structure-based approach while maintaining our ability to efficiently compute the value of each outcome. It does this by utilizing kernel-based methods, popular in machine learning (=-=Muller, Mika, Ratsch, Tsuda, & Scholkopf, 2001-=-). Of course, there is no entirely free lunch, and some other assumptions are hidden behind these technique—and we will try to understand them later on. The basic idea behind structure-free compilatio... |

323 | 1981: ‘The Framing of - Tversky, Kahneman |

310 | Intransitivity of Preferences - Tversky - 1969 |

279 |
Some Results on Tchebycheffian Spline Functions
- Kimeldorf, Wahba
- 1971
(Show Context)
Citation Context ... it turns out that these complexity issues can be overcome by using some duality techniques from optimization theory (Bertsekas, Nedic, & Ozdaglar, 2003) and Reproducing Kernel Hilbert Spaces (RKHS) (=-=Kimeldorf & Wahba, 1971-=-). The bottom line is that these techniques allow for computing a solution for C (corresponding to the desired value function V : F ↦→ ℜ), and subsequently computing the values of V on the outcomes Ω,... |

225 |
Statistical Decision Functions
- Wald
- 1950
(Show Context)
Citation Context ... Raynaud, 1986; Debreu, 1954; von Neumann & Morgenstern, 1947; Fishburn, 1969), in mathematics (Birkhoff, 1948; Davey & Priestley, 2002), in operations research and decision analysis (Fishburn, 1974; =-=Wald, 1950-=-; Bouyssou, Marchant, Pirlot, Tsoukias, & Vincke, 2006), in psychology (Tversky, 1967, 1969; Kahneman & Tversky, 1979, 1984), and in various disciplines of compute science: AI (Doyle & Thomason, 1999;... |

203 |
Axiomatic Derivation of the Principle of Maximum Entropy and the Principle of Minimum Crossentropy
- Shore, Johnson
- 1980
(Show Context)
Citation Context ...process in statistical machine learning, our inference process is closer to more general settings of inference via constraint optimization such as, e.g., inference about probability distributions in (=-=Shore & Johonson, 1980-=-). Figure 8 relates this discussion and the concepts involved to our meta-model. When we have a complete specification, the inference process leaves us with a single model, as shown in Figure 9. 24Fi... |

196 | Toward a Logic for Qualitative Decision Theory
- Boutilier
- 1994
(Show Context)
Citation Context ...imony, 2006). Preference languages and logics have been discussed by philosophers (Hansson, 2001a) and AI researchers (Brewka, Niemelä, & Truszczynski, 2003; Tan & Pearl, 1994; Doyle & Wellman, 1994; =-=Boutilier, 1994-=-). In fact, the best known interpretation for non-monotonic logics is known as the preference semantics (Shoham, 1987), and thus most nonmonotonic logics, such as (Kraus, Lehmann, & Magidor, 1990; McC... |

182 |
SemiringBased Constraint Solving and Optimization
- Bistarelli, Montanari, et al.
- 1997
(Show Context)
Citation Context ... slightly different perspective). Value functions map tuples to integers, or reals. More generally, mappings to other structures can be defined, as has been done in the framework of soft constraints (=-=Bistarelli, Montanari, & Rossi, 1997-=-; Bistarelli, Fargier, Montanari, Rossi, Schiex, & Verfaillie, 1999). This formalism can be viewed as generalizing the notion of GAI, as it allows for both new ranges for the function and aggregation ... |

173 |
Elimination by Aspects: A Theory of Choice
- Tversky
- 1972
(Show Context)
Citation Context ... algorithm. BIBLIOGRAPHIC NOTES Preference elicitation and value of information have been first studies in the areas of decision analysis and psychology, where it remains a topic of great importance (=-=Tversky, 1972-=-; Keeney & Raiffa, 1976; Howard & Matheson, 1984; French, 1986). This line of research has been extended in artificial intelligence, with a focus on automating the process of preference elicitation (H... |

153 |
Convex Analysis and Optimization; Athena Scientific
- Bertsekas, Nedić, et al.
- 2003
(Show Context)
Citation Context ... fact, of each individual constraint in C, can be exponential in n. Interestingly, it turns out that these complexity issues can be overcome by using some duality techniques from optimization theory (=-=Bertsekas, Nedic, & Ozdaglar, 2003-=-) and Reproducing Kernel Hilbert Spaces (RKHS) (Kimeldorf & Wahba, 1971). The bottom line is that these techniques allow for computing a solution for C (corresponding to the desired value function V :... |

144 |
Notes on the Theory of Choice
- Kreps
- 1988
(Show Context)
Citation Context ...s been a lot of work on this topic, especially by economists, who have attempted to come up with weaker or simply different properties (and there’s a very nice and accessible text book on this topic (=-=Kreps, 1988-=-)). Here, however, we’ll stick to the classical result. Formally, our model here has the following elements: • Ω – Set of possible concrete outcomes • L = Π(Ω) – Set of possible lotteries (that is, pr... |

124 |
A Framework for Expressing and Combining Preferences
- Agrawal, Wimmers
(Show Context)
Citation Context ...ot, Tsoukias, & Vincke, 2006), in psychology (Tversky, 1967, 1969; Kahneman & Tversky, 1979, 1984), and in various disciplines of compute science: AI (Doyle & Thomason, 1999; Doyle, 2004), databases (=-=Agrawal & Wimmers, 2000-=-; Kießling, 2002; Chomicki, 2002), HCI (Gajos & Weld, 2005; Chen & Pu, 2007), electronic commerce (Sandholm & Boutilier, 2006a), and more. 2. Modeling Preferences Two key questions arise when one appr... |

124 | Graphical Models for Preference and Utility - Bacchus, Grove - 1995 |

107 | A POMDP Formulation of Preference Elicitation Problems
- Boutilier
(Show Context)
Citation Context ...e elicitation for policy optimization as a properly defined decision process was first suggested in (Chajewska, Koller, & Parr, 2000; Chajewska, Getoor, Norman, & Shahar, 1998), and then extended in (=-=Boutilier, 2002-=-) which suggested the POMDP-based formulation. Preference elicitation under the minmax-regret model selection criterion has been studied in (Boutilier, Patrascu, Poupart, & Schuurmans, 2006; Braziunas... |

99 | H.: Semiring-based CSPs and valued CSPs: Frameworks, properties and comparison. Constraints 4
- Bistarelli, Montanari, et al.
- 1999
(Show Context)
Citation Context ...e functions map tuples to integers, or reals. More generally, mappings to other structures can be defined, as has been done in the framework of soft constraints (Bistarelli, Montanari, & Rossi, 1997; =-=Bistarelli, Fargier, Montanari, Rossi, Schiex, & Verfaillie, 1999-=-). This formalism can be viewed as generalizing the notion of GAI, as it allows for both new ranges for the function and aggregation and comparison operators other than addition and maximization. Valu... |

99 | Making Rational Decisions Using Adaptive Utility Elicitation
- Chajewska, Koller, et al.
(Show Context)
Citation Context ...iunas & Boutilier, 2005; Payne, Bettman, & Johnson, 1993; Smith & McGinty, 2003). Casting preference elicitation for policy optimization as a properly defined decision process was first suggested in (=-=Chajewska, Koller, & Parr, 2000-=-; Chajewska, Getoor, Norman, & Shahar, 1998), and then extended in (Boutilier, 2002) which suggested the POMDP-based formulation. Preference elicitation under the minmax-regret model selection criteri... |

99 | Preference Elicitation in Combinatorial Auctions
- Sandholm, Boutilier
- 2006
(Show Context)
Citation Context ...ines of compute science: AI (Doyle & Thomason, 1999; Doyle, 2004), databases (Agrawal & Wimmers, 2000; Kießling, 2002; Chomicki, 2002), HCI (Gajos & Weld, 2005; Chen & Pu, 2007), electronic commerce (=-=Sandholm & Boutilier, 2006-=-a), and more. 2. Modeling Preferences Two key questions arise when one approaches a modeling task, namely (1) What is the model? (2) What questions/queries do we want to ask about this model. 3To ach... |

98 |
A Semantics Approach to Nonmonotonic Logics
- Shoham
- 1987
(Show Context)
Citation Context ...rewka, Niemelä, & Truszczynski, 2003; Tan & Pearl, 1994; Doyle & Wellman, 1994; Boutilier, 1994). In fact, the best known interpretation for non-monotonic logics is known as the preference semantics (=-=Shoham, 1987-=-), and thus most nonmonotonic logics, such as (Kraus, Lehmann, & Magidor, 1990; McCarthy, 1980; Reiter, 1980), can be viewed as expressing preference statements, usually associated with aggressive int... |

85 | Foundations of measurement
- Krantz, Luce, et al.
- 1971
(Show Context)
Citation Context ...ics. BIBLIOGRAPHIC NOTES Work on preference models, algorithms, and representation appears in many areas. Discussion of preference models are abundant in the philosophical literature (Hansson, 2001a; =-=Krantz, Luce, Suppes, & Tversky, 1971-=-; von Wright, 1963; Halldén, 1957), in economics and game-theory (K.J.Arrow & Raynaud, 1986; Debreu, 1954; von Neumann & Morgenstern, 1947; Fishburn, 1969), in mathematics (Birkhoff, 1948; Davey & Pri... |

84 | Background to Qualitative Decision Theory
- Doyle, Thomason
- 1999
(Show Context)
Citation Context ...hburn, 1974; Wald, 1950; Bouyssou, Marchant, Pirlot, Tsoukias, & Vincke, 2006), in psychology (Tversky, 1967, 1969; Kahneman & Tversky, 1979, 1984), and in various disciplines of compute science: AI (=-=Doyle & Thomason, 1999-=-; Doyle, 2004), databases (Agrawal & Wimmers, 2000; Kießling, 2002; Chomicki, 2002), HCI (Gajos & Weld, 2005; Chen & Pu, 2007), electronic commerce (Sandholm & Boutilier, 2006a), and more. 2. Modeling... |

79 |
Representation of a Preference Ordering By A Numerical Function
- Debreu
- 1954
(Show Context)
Citation Context ... models are abundant in the philosophical literature (Hansson, 2001a; Krantz, Luce, Suppes, & Tversky, 1971; von Wright, 1963; Halldén, 1957), in economics and game-theory (K.J.Arrow & Raynaud, 1986; =-=Debreu, 1954-=-; von Neumann & Morgenstern, 1947; Fishburn, 1969), in mathematics (Birkhoff, 1948; Davey & Priestley, 2002), in operations research and decision analysis (Fishburn, 1974; Wald, 1950; Bouyssou, Marcha... |

79 |
The Foundations of Expected Utility
- Fishburn
- 1982
(Show Context)
Citation Context ...ames and Economic Behavior (von Neumann & Morgenstern, 1947). Since then, many economists have tried to provide similar axiomatizations of choice behavior and this topic is discussed by (Kreps, 1988; =-=Fishburn, 1982-=-). One of the most beautiful results in this area is presented in Savage’s The Foundation of Statistics (Savage, 1972). 3. Preference Elicitation Our discussion so far has focused on models and langua... |

65 | Representing Preferences as Ceteris Paribus Comparatives
- Doyle, Wellman
- 1994
(Show Context)
Citation Context ...Brafman, Domshlak, & Shimony, 2006). Preference languages and logics have been discussed by philosophers (Hansson, 2001a) and AI researchers (Brewka, Niemelä, & Truszczynski, 2003; Tan & Pearl, 1994; =-=Doyle & Wellman, 1994-=-; Boutilier, 1994). In fact, the best known interpretation for non-monotonic logics is known as the preference semantics (Shoham, 1987), and thus most nonmonotonic logics, such as (Kraus, Lehmann, & M... |

61 | A Family of Additive Online Algorithms for Category Ranking
- Crammer, Singer
- 2003
(Show Context)
Citation Context ...Domshlak, 2008) Much recent work in the machine learning community is devoted to rank learning problems, where we attempt to learn a ranking function on some input based on examples. See for example (=-=Crammer & Singer, 2003-=-; Radlinski & Joachims, 2007; Burges, Shaked, Renshaw, Lazier, Deeds, Hamilton, & Hullender, 2005). 2.4 Uncertainty and Utility Functions In our discussion so far, we assumed that our preferences were... |

61 |
Qualitative Decision Theory
- Tan, Pearl
- 1994
(Show Context)
Citation Context ...i, & Wilson, 2005; Brafman, Domshlak, & Shimony, 2006). Preference languages and logics have been discussed by philosophers (Hansson, 2001a) and AI researchers (Brewka, Niemelä, & Truszczynski, 2003; =-=Tan & Pearl, 1994-=-; Doyle & Wellman, 1994; Boutilier, 1994). In fact, the best known interpretation for non-monotonic logics is known as the preference semantics (Shoham, 1987), and thus most nonmonotonic logics, such ... |

60 |
Decision Theory
- French
- 1986
(Show Context)
Citation Context ...e of information have been first studies in the areas of decision analysis and psychology, where it remains a topic of great importance (Tversky, 1972; Keeney & Raiffa, 1976; Howard & Matheson, 1984; =-=French, 1986-=-). This line of research has been extended in artificial intelligence, with a focus on automating the process of preference elicitation (Ha & Haddawy, 1997, 1999; Torrens, Faltings, & Pu, 2002; Pu & F... |

57 |
Lexicographic Orders, Utilities, and Decision Rules: A Survey. Management Science 20(11): 1442–1471
- Fishburn
- 1974
(Show Context)
Citation Context ...ory (K.J.Arrow & Raynaud, 1986; Debreu, 1954; von Neumann & Morgenstern, 1947; Fishburn, 1969), in mathematics (Birkhoff, 1948; Davey & Priestley, 2002), in operations research and decision analysis (=-=Fishburn, 1974-=-; Wald, 1950; Bouyssou, Marchant, Pirlot, Tsoukias, & Vincke, 2006), in psychology (Tversky, 1967, 1969; Kahneman & Tversky, 1979, 1984), and in various disciplines of compute science: AI (Doyle & Tho... |

56 | Active Exploration for Learning Rankings From Clickthrough Data
- Radlinski, Joachims
(Show Context)
Citation Context ...ent work in the machine learning community is devoted to rank learning problems, where we attempt to learn a ranking function on some input based on examples. See for example (Crammer & Singer, 2003; =-=Radlinski & Joachims, 2007-=-; Burges, Shaked, Renshaw, Lazier, Deeds, Hamilton, & Hullender, 2005). 2.4 Uncertainty and Utility Functions In our discussion so far, we assumed that our preferences were over outcomes that we can o... |

55 | Preference-Based Constrained Optimization with Cp-Nets
- Boutilier, Brafman, et al.
(Show Context)
Citation Context ...used. Specifically, we consider the language of statements of preference on the values of single attributes. This specific language underlies the preference specification approach called CP-networks (=-=Boutilier, Brafman, Domshlak, Hoos, & Poole, 2004-=-a), and the components of this meta-model instance are described in Figure 4. The model of user preferences assumed in the CP-networks approach is that of partial orders over outcomes. The reason is p... |

55 | SmartClients: Constraint Satisfaction as a Paradigm for Scaleable Intelligent Information Systems. Constraints 7(1
- Torrens, Faltings, et al.
- 2002
(Show Context)
Citation Context ...Howard & Matheson, 1984; French, 1986). This line of research has been extended in artificial intelligence, with a focus on automating the process of preference elicitation (Ha & Haddawy, 1997, 1999; =-=Torrens, Faltings, & Pu, 2002-=-; Pu & Faltings, 2004; Faltings, Torrens, & Pu, 2004; Braziunas & Boutilier, 2005; Payne, Bettman, & Johnson, 1993; Smith & McGinty, 2003). Casting preference elicitation for policy optimization as a ... |

53 | Utility Elicitation As A Classification Problem
- Chajewska, Getoor, et al.
- 1998
(Show Context)
Citation Context ...ettman, & Johnson, 1993; Smith & McGinty, 2003). Casting preference elicitation for policy optimization as a properly defined decision process was first suggested in (Chajewska, Koller, & Parr, 2000; =-=Chajewska, Getoor, Norman, & Shahar, 1998-=-), and then extended in (Boutilier, 2002) which suggested the POMDP-based formulation. Preference elicitation under the minmax-regret model selection criterion has been studied in (Boutilier, Patrascu... |

53 | Prospects for Preferences
- Doyle
- 2004
(Show Context)
Citation Context ... Bouyssou, Marchant, Pirlot, Tsoukias, & Vincke, 2006), in psychology (Tversky, 1967, 1969; Kahneman & Tversky, 1979, 1984), and in various disciplines of compute science: AI (Doyle & Thomason, 1999; =-=Doyle, 2004-=-), databases (Agrawal & Wimmers, 2000; Kießling, 2002; Chomicki, 2002), HCI (Gajos & Weld, 2005; Chen & Pu, 2007), electronic commerce (Sandholm & Boutilier, 2006a), and more. 2. Modeling Preferences ... |

52 | Social Learning Theory
- unknown authors
- 1977
(Show Context)
Citation Context ...f, 1948; Davey & Priestley, 2002), in operations research and decision analysis (Fishburn, 1974; Wald, 1950; Bouyssou, Marchant, Pirlot, Tsoukias, & Vincke, 2006), in psychology (Tversky, 1967, 1969; =-=Kahneman & Tversky, 1979-=-, 1984), and in various disciplines of compute science: AI (Doyle & Thomason, 1999; Doyle, 2004), databases (Agrawal & Wimmers, 2000; Kießling, 2002; Chomicki, 2002), HCI (Gajos & Weld, 2005; Chen & P... |

52 |
mCP Nets: Representing and Reasoning with Preferences of Multiple Agents
- Rossi, Venable, et al.
- 2004
(Show Context)
Citation Context ...OTES CP-nets were introduced in (Boutilier et al., 2004a; Boutilier, Brafman, Domshlak, Hoos, & Poole, 2004b) and have generated various extensions, enhancements, and analysis, such as (Wilson, 2004; =-=Rossi, Venable, & Walsh, 2004-=-; Gold4. When the CP-net is fully specified, i.e., an ordering over the domain of each attribute is specified for every possible assignment to the parents, we know that a single most preferred assignm... |

49 | On Graphical Modeling of Preference and Importance
- Brafman, Domshlak, et al.
- 2006
(Show Context)
Citation Context ...g the feasible assignments, then a number of Paretooptimal assignments may exist – i.e., assignments o such that for o ⊀ o ′ for any other feasible o ′ . 15smith, Lang, Truszczynski, & Wilson, 2005; =-=Brafman, Domshlak, & Shimony, 2006-=-). Preference languages and logics have been discussed by philosophers (Hansson, 2001a) and AI researchers (Brewka, Niemelä, & Truszczynski, 2003; Tan & Pearl, 1994; Doyle & Wellman, 1994; Boutilier, ... |

49 |
Conjoint Measurement for Quantifying Judgmental Data
- Green, Rao
- 1971
(Show Context)
Citation Context ...tion in ℜ4n. 6 BIBLIOGRAPHIC NOTES The idea of finding factored value functions that represent some preference relation is discussed at length in (Keeney & Raiffa, 1976). Conjoint measurement theory (=-=Green & Rao, 1971-=-; Green, Krieger, & Wind, 2001) specifically considers the problem of generating a value function describing a user (or a populations) preference based on answers to preference queries. Most of this w... |

42 | GAI Networks for Utility Elicitation
- Gonzales, Perny
(Show Context)
Citation Context ...preferences of the user. However, computing the optimal outcome, that is, finding the value of X that maximizes V is more difficult. To deal with such queries, it is useful to maintain a GAI-network (=-=Gonzales & Perny, 2004-=-). A GAInetwork is an annotated graph whose nodes correspond to the attributes in X. An edge connects the nodes corresponding to a pair of attributes Xi, Xj ∈ X if Xi and Xj occur jointly in some fact... |

42 | Problem-Focused Incremental Elicitation of Multiattribute Utility Models
- Ha, Haddawy
- 1997
(Show Context)
Citation Context ...2; Keeney & Raiffa, 1976; Howard & Matheson, 1984; French, 1986). This line of research has been extended in artificial intelligence, with a focus on automating the process of preference elicitation (=-=Ha & Haddawy, 1997-=-, 1999; Torrens, Faltings, & Pu, 2002; Pu & Faltings, 2004; Faltings, Torrens, & Pu, 2004; Braziunas & Boutilier, 2005; Payne, Bettman, & Johnson, 1993; Smith & McGinty, 2003). Casting preference elic... |

42 | Utility Functions for Ceteris Paribus Preferences
- McGeachie, Doyle
(Show Context)
Citation Context ... based on answers to preference queries. Most of this work seeks an additive independent model, though some extensions exist (Bouyssou & Pirlot, 2005). Compilation methods were investigated in AI by (=-=McGeachie & Doyle, 2004-=-; Domshlak & Joachims, 2007; Brafman & Domshlak, 2008) Much recent work in the machine learning community is devoted to rank learning problems, where we attempt to learn a ranking function on some inp... |

41 | Preference Elicitation for Interface Optimization
- Gajos, Weld
(Show Context)
Citation Context ...69; Kahneman & Tversky, 1979, 1984), and in various disciplines of compute science: AI (Doyle & Thomason, 1999; Doyle, 2004), databases (Agrawal & Wimmers, 2000; Kießling, 2002; Chomicki, 2002), HCI (=-=Gajos & Weld, 2005-=-; Chen & Pu, 2007), electronic commerce (Sandholm & Boutilier, 2006a), and more. 2. Modeling Preferences Two key questions arise when one approaches a modeling task, namely (1) What is the model? (2) ... |

41 |
Social Choice and Multicriterion Decision Making
- Arrow, Raynaud
- 1986
(Show Context)
Citation Context ...iscussion of preference models are abundant in the philosophical literature (Hansson, 2001a; Krantz, Luce, Suppes, & Tversky, 1971; von Wright, 1963; Halldén, 1957), in economics and game-theory (K.J.=-=Arrow & Raynaud, 1986-=-; Debreu, 1954; von Neumann & Morgenstern, 1947; Fishburn, 1969), in mathematics (Birkhoff, 1948; Davey & Priestley, 2002), in operations research and decision analysis (Fishburn, 1974; Wald, 1950; Bo... |

40 | Preference Logic - Hansson |