1
Preference Elicitation in Single and Multiple User Settings
Darius Braziunas, Craig Boutilier, 2005(Boutilier, Patrascu, Poupart, Shuurmans, 2003, 2005)
Nathanael Hyafil, Craig Boutilier, 2006a, 2006b
Department of Computer Science
University of Toronto
2
Overview
Preference Elicitation in A.I.
Single User Elicitation
• Foundations of Local queries [BB-05]
• Bayesian Elicitation [BB-05]
• Regret-based Elicitation [BPPS-03,05]
Multi-agent Elicitation (Mechanism Design)
• One-Shot Elicitation [HB-06b]
• Sequential Mechanisms [HB-06a]
3
Preference Elicitation in AI
Luggage Capacity?Two Door? Cost?
Engine Size?Color? Options?
Shopping for a Car:
4
The Preference Bottleneck
Preference elicitation: the process of determining a user’s preferences/utilities to the extent necessary to make a decision on her behalf
Why a bottleneck?• preferences vary widely
• large (multiattribute) outcome spaces
• quantitative utilities (the “numbers”) difficult to assess
5
Automated Preference Elicitation
The interesting questions:• decomposition of preferences
• what preference info is relevant to the task at hand?
• when is the elicitation effort worth the improvement it offers in terms of decision quality?
• what decision criterion to use given partial utility info?
6
Overview
Preference Elicitation in A.I.
• Constraint-based Optimization
• Factored Utility Models
• Types of Uncertainty
• Types of Queries
Single User Elicitation
Multi-agent Elicitation (Mechanism Design)
7
Constraint-based Decision Problems
Constraint-based optimization (CBO):
• outcomes over variables X = {X1 … Xn}
• constraints C over X spell out feasible decisions
• generally compact structure, e.g., X1 & X2 ¬
X3
• add a utility function u: Dom(X) → R
• preferences over configurations
8
Constraint-based Decision Problems
Must express u compactly like C• generalized additive independence (GAI)
model proposed by Fishburn (1967) [and BG95]
nice generalization of additive linear models
• given by graphical model capturing independence
9
Factored Utilities: GAI ModelsSet of K factors fk over subset of vars X[k]
• “local” utility for each local configuration
•
[Fishburn67] u in this form exists iff• lotteries p and q are equally preferred whenever p and q
have the same marginals over each X[k]
Kkk kfu )xx) ][(( A Bf1(A)
a: 3a: 1
f2(B)b: 3b: 1
Cf3(BC)
bc: 12 bc: 2
…u(abc) = f1(a)+ f2(b)+ f3(bc)
10
Optimization with GAI Models
Optimize using simple IP (or Var Elim, or…)• number of vars linear in size of GAI model
Kk
k kfu )xx) ][((
A Bf1(A)a: 3a: 1
f2(B)b: 3b: 1
C f3(BC) bc: 12 bc: 2
…
CAIukDomkkik
kkXI , tosubj. max
])[(][][][][},{
Xxxxx
11
Difficulties in CBO
Constraint elicitation• generally stable across different users
Utility (objective) representation• GAI solidifies many of intuitions in soft constraints
Utility elicitation: how do we assess individual user preferences?
• need to elicit GAI model structure (independence)
• need to elicit (constraints on) GAI parameters
• need to make decisions with imprecise parameters
12
Strict Utility Function Uncertainty
User’s actual utility u unknown
Assume feasible set F U = [0,1]n
• allows for unquantified or “strict” uncertainty
• e.g., F a set of linear constraints on GAI terms
How should one make a decision? elicit info?
u(red,2door,280hp) > 0.4u(red,2door,280hp) > u(blue,2door,280hp)
13
f2(L,N)l,n: [2,4]l,n: [1,2]
Strict Uncertainty Representation
L
NP
Utility Function
f1(L)l: [7,11]l: [2,5]
14
Bayesian Utility Function Uncertainty
User’s actual utility u unknown
Assume density P over U = [0,1]n
Given belief state P, EU of decision x is:
EU(x , P) = U (px . u) P( u ) du
Decision making is easy, but elicitation harder?• must assess expected value of information in query
15
f2(L,N)l,n: l,n:
…
Bayesian Representation
L
NP
Utility Function
f1(L)l: l:
16
Query Types
Comparison queries (is x preferred to x’ ?)• impose linear constraints on parameters
k fk(x[k]) > k fk(x’[k])
• Interpretation is straightforward
U
17
Query Types
Bound queries (is fk(x[k]) > v ?)
• response tightens bound on specific utility parameter
• can be phrased as a local standard gamble query
U
18
Overview
Preference Elicitation in A.I.
Single User Elicitation
• Foundations of Local queries
• Bayesian Elicitation
• Regret-based Elicitation
Multi-agent Elicitation (Mechanism Design)
• One-Shot Elicitation
• Sequential Mechanisms
19
Difficulties with Bound Queries
Bound queries focus on local factors• but values cannot be fixed without reference to others!
• seemingly “different” local prefs correspond to same u
u(Color,Doors,Power) = u1(Color,Doors) +
u2(Doors,Power)
u(red,2door,280hp) = u1(red,2door) +
u2(2door,280hp)
u(red,4door,280hp) = u1(red,4door) +
u2(4door,280hp)
10 6 4
6 3 3
1 9
20
Local Queries [BB05]
We wish to avoid queries on whole outcomes• can’t ask purely local outcomes
• but can condition on a subset of default values
Conditioning set C(f) for factor fi(Xi) :
• variables that share factors with Xi
• setting default outcomes on C(f) renders Xi independent of remaining variables
• enables local calibration of factor values
21
Local Standard Gamble Queries
Local std. gamble queries• use “best” and “worst” (anchor) local outcomes
-- conditioned on default values of conditioning set
• bound queries on other parameters relative to these
• gives local value function v(x[i]) (e.g., v(ABC) )
Hence we can legitimately ask local queries:
But local Value Functions not enough: • must calibrate: requires global scaling
22
Global Scaling
Elicit utilities of anchor outcomes wrt global
best and worst outcomes
• the 2*m “best” and “worst” outcomes for each
factor
• these require global std gamble queries
(note: same is true for pure additive models)
23
Bound Query Strategies
Identify conditioning sets Ci for each factor fiDecide on “default” outcome
For each fi identify top and bottom anchors
• e.g., the most and least preferred values of factor i
• given default values of Ci
Queries available:• local std gambles: use anchors for each factor, C-sets
• global std gambles: gives bounds on anchor utilities
24
Overview
Preference Elicitation in A.I.
Single User Elicitation
• Foundations of Local queries
• Bayesian Elicitation
• Regret-based Elicitation
Multi-agent Elicitation (Mechanism Design)
• One-Shot Elicitation
• Sequential Mechanisms
25
Partial preference informationBayesian uncertainty
Probability distribution p over utility functions Maximize expected (expected) utility
MEU decision x* = arg maxx Ep [u(x)]
Consider:• elicitation costs
• values of possible decisions
• optimal tradeoffs between elicitation effort and improvement in decision quality
26
Query selection
At each step of elicitation process, we can• obtain more preference information
• make or recommend a terminal decision
27
Bayesian approachMyopic EVOI
MEU(p)
q1q2
...
r1,1 r2,1r1,2 r2,2
...
MEU(p1,1) MEU(p1,2) MEU(p2,1) MEU(p2,2)
28
Expected value of information
MEU(p) = Ep [u(x*)]
Expected posterior utility: EPU(q,p) = Er|q,p [MEU(pr)]
Expected value of information of query q:
EVOI(q) = EPU(q,p) – MEU(p)
MEU(p)
q1q2 ...
r1,1 r2,1r1,2 r2,2
...MEU(p1,1) MEU(p1,2) MEU(p2,1) MEU(p2,2)
29
Bayesian approachMyopic EVOI
Ask query with highest EVOI - cost [Chajewska et al ’00]
• Global standard gamble queries (SGQ) “Is u(oi) > l?”
• Multivariate Gaussian distributions over utilities
[Braziunas and Boutilier ’05]• Local SGQ over utility factors
• Mixture of uniforms distributions over utilities
30
Local elicitation in GAI models [Braziunas and Boutilier ’05]
Local elicitation procedure• Bayesian uncertainty over local factors• Myopic EVOI query selection
Local comparison query
“Is local value of factor setting xi greater than l”?
• Binary comparison query
• Requires yes/no response
• query point l can be optimized analytically
31
Experiments
Car rental domain: 378 parameters [Boutilier et al. ’03]
• 26 variables, 2-9 values each, 13 factors
2 strategies• Semi-random query
Query factor and local configuration chosen at randomQuery point set to the mean of local value function
• EVOI querySearch through factors and local configurationsQuery point optimized analytically
32
Experiments
0 10 20 30 40 50 60 70 80 90 1000
2
4
6
8
10
12
14
16
Mixture of Uniforms
Uniform
Gaussian
No. of queries
Percentage utility error (w.r.t. truemax utility)
33
Bayesian Elicitation: Future Work
GAI structure elicitation and verification
Sequential EVOI
Noisy responses
34
Overview
Single User Elicitation
• Foundations of Local queries
• Bayesian Elicitation
• Regret-based Elicitation why MiniMax Regret (MMR) ?
Decision making with MMR
Elicitation with MMR
Multi-agent Elicitation (Mechanism Design)
35
Minimax Regret: Utility Uncertainty
Regret of x w.r.t. u:
Max regret of x w.r.t. F:
Decision with minimax regret w.r.t. F:
),(),(),( * uxEUuxEUuxR u
),(max),( uxRFxMRFu
),()(;),(minarg *
)(
* FxMRFMMRFxMRx FXFeasx
F
36
Why Minimax Regret?*
Appealing decision criterion for strict uncertainty• contrast maximin, etc.
• not often used for utility uncertainty [BBB01,HS010]
x
x’
x’x
x
x’x’
x
xx’
x
x’
Bett
er
u1 u2 u3 u4 u5 u6
37
Why Minimax Regret?
Minimizes regret in presence of adversary• provides bound worst-case loss
• robustness in the face of utility function uncertainty
In contrast to Bayesian methods:• useful when priors not readily available
• can be more tractable; see [CKP00/02, Bou02]
• effective elicitation even if priors available [WB03]
38
Overview
Single User Elicitation
• Foundations of Local queries
• Bayesian Elicitation
• Regret-based Elicitation why MiniMax Regret (MMR) ?
Decision making with MMR
Elicitation with MMR
Multi-agent Elicitation (Mechanism Design
39
Computing Max Regret
Max regret MR(x,F) computed as an IP• number of vars linear in GAI model size
• number of (precomputed) constants (i.e., local regret terms) quadratic in GAI model size
• r( x[k] , x’[k] ) = u(x’[k] ) – u(x[k] )
CAIrk k
kkkXI ik
, tosubj. max]['
][']['][}',{ ][' x
xxxx
40
Minimax Regret in Graphical Models
We convert minimax to min (standard trick)• obtain a MIP with one constraint per feasible config
• linearly many vars (in utility model size)
Key question: can we avoid enumerating all x’ ?
41
Constraint Generation
Very few constraints will be active in solution
Iterative approach: • solve relaxed IP (using a subset of
constraints)
• Solve for maximally violated constraint
• if any add it and repeat; else terminate
42
Constraint Generation Performance
Key properties:• aim: graphical structure permits practical
solution
• convergence (usually very fast, few constraints)
• very nice anytime properties
• considerable scope for approximation
• produces solution x* as well as witness xw
43
Overview
Single User Elicitation
• Foundations of Local queries
• Bayesian Elicitation
• Regret-based Elicitation why MiniMax Regret (MMR) ?
Decision making with MMR
Elicitation with MMR
Multi-agent Elicitation (Mechanism Design)
44
Regret-based Elicitation[Boutilier, Patrascu, Poupart, Schuurmans IJCAI05; AIJ 06]
Minimax optimal solution may not be satisfactoryImprove quality by asking queries
• new bounds on utility model parameters
Which queries to ask?• what will reduce regret most quickly?
• myopically? sequentially?
Closed form solution seems infeasible• to date we’ve looked only at heuristic elicitation
45
Elicitation Strategies IHalve Largest Gap (HLG)
• ask if parameter with largest gap > midpoint
• MMR(U) ≤ maxgap(U), hence nlog(maxgap(U)/) queries needed to reduce regret to
• bound is tight
• like polyhedral-based conjoint analysis [THS03]
f1(a,b) f1(a,b) f1(a,b) f1(a,b) f2(b,c) f2(b,c) f2(b,c) f2(b,c)
46
Elicitation Strategies II
Current Solution (CS)• only ask about parameters of optimal solution x* or
regret-maximizing witness xw
• intuition: focus on parameters that contribute to regret reducing u.b. on xw or increasing l.b. on x* helps
• use early stopping to get regret bounds (CS-5sec)
f1(a,b) f1(a,b) f1(a,b) f1(a,b) f2(b,c) f2(b,c) f2(b,c) f2(b,c)
47
Elicitation Strategies IIIOptimistic-pessimistic (OP)
• query largest-gap parameter in one of:optimistic solution xo
pessimistic solution xp
Computation:• CS needs minimax optimization• OP needs standard optimization• HLG needs no optimization
Termination:• CS easy• Others ?
48
Results (Small Random)
10vars; < 5 vals
10 factors, at most 3 vars
Avg 45 trials
49
Results (Car Rental, Unif)
26 vars; 61 billion configs
36 factors, at most 5 vars; 150 parameters
Avg 45 trials
50
Results (Real Estate, Unif)
20 vars; 47 million configs
29 factors, at most 5 vars; 100 parameters
Avg 45 trials
51
Results (Large Rand, Unif)
25 vars; < 5 vals
20 factors, at most 3 vars
A 45 trials
52
Summary of Results
CS works best on test problems• time bounds (CS-5): little impact on query quality
• always know max regret (or bound) on solution
• time bound adjustable (use bounds, not time)
OP competitive on most problems• computationally faster (e.g., 0.1s vs 14s on RealEst)
• no regret computed so termination decisions harder
HLG much less promising
53
Interpretation
HLG:
• provable regret reduced very quickly
But:
• true regret faster (often to optimality)
• OP and CS restricted to feasible decisions
• CS focuses on relevant parameters
54
Conclusion – Single User
Local parameter elicitation• Theoretically sound
• Computationally practical
• Easier to answer
Bayesian EVOI / Regret-based elicitation• Good guides for elicitation
• Integrated in computationally tractable algorithms
Future Work:• Sequential reasoning
55
Questions?
References:D. Braziunas and C. Boutilier:
• “Local Utility Elicitation in GAI Models”, UAI 2005
C. Boutilier, R. Patrascu, P. Poupart, D.
Shuurmans:
• “Constraint-based Optimization and Utility Elicitation
using the Minimax Decision Criterion”, Artificial
Intelligence, 2006
• (CP-2003 + IJCAI 2005)
56
Preference Elicitation in
Single and Multiple User Settings
Part 2
Darius Braziunas, Craig Boutilier, 2005(Boutilier, Patrascu, Poupart, Shuurmans, 2003, 2005)
Nathanael Hyafil, Craig Boutilier, 2006a, 2006b
Department of Computer Science
University of Toronto
57
Overview
Single User Elicitation• Foundations of Local queries
• Bayesian Elicitation
• Regret-based Elicitation
Multi-agent Elicitation
• Background: Mechanism Design
• Partial Revelation Mechanisms
• One-Shot Elicitation
• Sequential Mechanisms
58
Bargaining for a Car
Luggage Capacity?Two Door? Cost?
Engine Size?Color? Options?
$$$$
$$
$$
$$
$$
$$
$$
59
Multiagent PE: Mechanism Design
Incentive to misrepresent preferencesMechanism design tackles this:
• Design rules of game to induce behavior that leads to maximization of some objective
(e.g., Social Welfare, Revenue, ...)
• Objective value depends on private information held by self-interested agents Elicitation + Incentives
Applications:• Auctions, multi-attribute Negotiation, Procurement problems,
• Network protocols, Autonomic computing, ...
60
Basic Social Choice Setup
Choice of x from outcomes X
Agents 1..n: type ti Ti and valuation vi(x, ti)
Type vectors: tT and t-i T-i
Goal: optimize social choice function f: T X• e.g., social welfare SW(x,t) = vi(x, ti)
Assume payments and quasi-linear utility:• ui(x, i ,ti ) = vi(x, ti ) - i
Our focus: SW maximization, quasi-linear utility
61
Basic Mechanism Design
A mechanism m consists of three components: • actions Ai
• allocation function O: A X
• payment functions pi : A R
m induces a Bayesian game• m implements social choice function f if
in equilibrium : O((t)) = f(t) for all tT
62
Incentive Compatibility (Truth-telling)
Dominant Strategy IC
• No matter what: agent i should tell the truth
Bayes-Nash IC
• Assume others tell the truth
• Assume agent i has Bayesian prior over others’ types
• Then, in expectation, agent i should tell the truth
Ex-Post IC
• Assume others tell the truth
• Assume agent i knows the others’ types
• Then agent i should tell the truth
63
Properties
Mechanism is Efficient: • maximizes SW given reported types:
-efficient: within of optimal SW
Ex Post Individually Rational:• No agent can lose by participating
-IR: can lose at most
64
Direct Mechanisms
Revelation principle: focus on direct mechanisms where agents directly and (in eq.) truthfully reveal their full types
For example, Groves scheme (e.g., VCG):• choose efficient allocation and use payment function:
• implements SWM in dominant strategies
• incentive compatible, efficient, individually rational
65
Groves Schemes
Strong results: Groves is basically the “only choice” for dominant strategy implementation
• Roberts (1979): only social choice functions implementable in dominant strategies are affine welfare maximizers (if all valuations possible)
• Green and Laffont (1977): must use Groves payments to implement affine maximizers
Implications for partial revelation
66
Issues with Classical Mechanism Design
Computation Costs
• e.g., Winner Determination
Revelation Costs
• Communication
• Computation
• Cognitive
• Privacy
67
Issues with Classical Mechanism Design
Full Revelation and “Quality”
• trade-off revelation costs with Social Welfare
Full Revelation and Incentives
• very dependent
• need “new” concepts
68
Overview
Single User Elicitation• Foundations of Local queries
• Bayesian Elicitation
• Regret-based Elicitation
Multi-agent Elicitation
• Background: Mechanism Design
• Partial Revelation Mechanisms
• One-Shot Elicitation
• Sequential Mechanisms
69
Partial Revelation Mechanisms
Full revelation unappealing
A partial type is any subset i Ti
A one-shot (direct) partial revelation mechanism• each agent reports a partial type i i
• typically i partitions type space, but not required
A truthful strategy: report i s.t. ti i
Goal: minimize revelation, computation, communication by suitable choice of partial types
70
Implications of Roberts
Partial revelation means we can’t generally maximize social welfare
• must allocate under type uncertainty
But if SCF is not an affine maximizer, we can’t expect dominant strategy implementation
What are some solutions?• relax solution concept to BNE / Ex-Post
• relax solution concept to approx incentives
• incremental and “hope for” less than full elicitation
• relax conditions on Roberts results
71
Existing Work on PRMs [Conen,Hudson,Sandholm, Parkes, Nisan&Segal, Blumrosen&Nisan]
Most Approaches:• require enough revelation to determine optimal
allocation and VCG paymentshence can’t offer savings in general [Nisan&Segal05]
• Sequential, not one-shot
• specific settings (1-item, combinatorial auctions)
Priority games [Blumrosen&Nisan 02]
• genuinely partial and approximate efficiency
• but very restricted valuation space (1-item)
72
Preference Elicitation in MechDes
We move beyond this by allowing approximately
optimal allocation• specifically, regret-based allocation models
Avoid Roberts by relaxing solution concept:• Bayes-Nash equilibrium?
NO! [HB-06b]
• Ex-Post IC?
NO ! [HB-06b]
• approximate, ex-post implementation
73
Partial Revelation MD:Impossibility Results
Bayes-Nash Equilibrium• Theorem: [HB-06b]
Deterministic PRMs are TrivialRandomized PRMs are Pseudo-Trivial
• Consequences: max expected SW = same as best trivialmax expected revenue = same as best trivial
“Useless”
Ex-Post Equilibrium• Same
74
Overview
Single User Elicitation• Foundations of Local queries
• Bayesian Elicitation
• Regret-based Elicitation
Multi-agent Elicitation
• Background: Mechanism Design
• Partial Revelation Mechanisms
• One-Shot Elicitation
• Sequential Mechanisms
75
Regret-based PRMs
In any PRM, how is allocation to be chosen?
x*() is minimax optimal decision for
A regret-based PRM: O()=x*() for all
76
Regret-based PRMs: Efficiency
Efficiency not possible with PRMs (unless MR=0)
• but bounds are quite obvious
Prop: If MR(x*(),) for all , then
regret-based PRM m is -efficient for truthtelling agents.
• thus we can tradeoff efficiency for elicitation effort
77
Regret-based PRMs: Incentives
Can generalize Groves payments• let fi (i) be an arbitrary type in i
Thm: Let m be a regret-based PRM with • partial types and a
• partial Groves payment scheme.
If MR(x*(),) for all , then m is
-ex post incentive compatible
78
Regret-based PRMs: Rationality
Can generalize Clark payments as well
Thm: Let m be a regret-based PRM with • partial types and a • partial Clark payment scheme.
If MR(x*(),) for all , then m is -ex post individually
rational.
A Clark-style regret-based PRM gives approximate efficiency, approximate IC (ex post) and approximate IR (ex post)
79
Approximate Incentives and IR
Natural to trade off efficiency for elicitation effortIs approximate IC acceptable?
• computing a good “lie”? Good? Huge computation costs
• if incentive to deviate from truth is small enough, then formal, approximate IC ensures practical, exact IC
Is approximate IR acceptable?• Similar argument
Thus regret-based PRMs offer scope to tradeoff• as long as we can find a good set of partial types
80
Computation and Design/Elicitation
Minimax optimization given partial type vector • same techniques as for single agent
• varies with setting (experiments: CBO with GAI)
Designing the mechanism• one-shot PRM: must choose partial types for each i
• sequential PRM: need elicitation strategy
• we apply generalization of CS to each task
81
(One-shot) Partial Type Optimization
Designing PRM: must pick partial types• we focus on bounds on utility parameters
A simple greedy approach• Let be current partial type vectors (initially {T} )
• Let =(1,… i,…n ) be partial type vector with greatest MMR
• Choose agent i and suitable split of partial type i into ’i and ’’i
• Replace all [i ] by pair of vectors: i ’i ;’’i
• Repeat until bound is acceptable
82
The Mechanism Tree
(’1,… i,…n ) (’’1,… i,…n )
(’1,… ’i,… ) (’1,… ’’i,… ) (’’1,… ’i,… ) (’’1,… ’’i,… )
(1,… i,…n )
83
A More Refined Approach
Simple model has drawbacks• exponential blowup (“naïve” partitioning)
• split of i useful in reducing regret in one partial type vector , but is applied at all partial type vectors
Refinement• apply split only at leaves where it is “useful”
keeps tree from blowing up, saves computation
• new splits traded off against “cached” splits
• once done, use either naïve/variable resolution types for each agent
84
Naïve vs. Variable Resolution
i
p1
p2
i
p1
p2
85
Heuristic for Choosing Splits
Adopt variant of current solution strategy
Let be partial type vector with max MMR• optimal solution x* regret-maximizing witness xw
• only split on parameters of utility functions of optimal solution x* or regret-maximizing witness xw
• intuition: focus on parameters that contribute to regret reducing u.b. on xw or increasing l.b. on x* helps
• pick agent-parameter pair with largest gap
86
Preliminary Empirical Results
Very preliminary results• use only very naïve algorithm
• single buyer, single seller
• 16 goods specified by 4 boolean variables
• valuation/cost given by GAI modeltwo factors, two vars each (buyer/seller factors are different)
thus 16 values/costs specified by 8 parametersno constraints on feasible allocations
87
Preliminary Empirical Results
88
Overview
Single User Elicitation• Foundations of Local queries
• Bayesian Elicitation
• Regret-based Elicitation
Multi-agent Elicitation
• Background: Mechanism Design
• Partial Revelation Mechanisms
• One-Shot Elicitation
• Sequential Mechanisms
89
Sequential PRMs
Optimization of one-shot PRMs unable to exploit conditional “queries”
• e.g., if seller cost of x greater than your upper bound, needn’t ask you for your valuation of x
Sequential PRMs• incrementally elicit partial type information
• apply similar heuristics for designing query policy
• incentive properties somewhat weaker: opportunity to manipulate payments by altering the query path
thus additional criteria can be used to optimize
90
Sequential PRMs: Definition
Set of queries Qi
• response rRi(qi) interpreted as partial type i (r) Ti
• history h: sequence of query-response pairs possibly followed by allocation (terminal)
Sequential mechanism m maps:• nonterminal histories to queries/allocations
• terminal histories to set of payment functions pi
Revealed partial type i(h): intersect. i (r), r in h
m is partial revelation if exists realizable terminal h s.t. i(h) admits more than one type ti
91
Sequential PRMs: Properties
Strategies i(hi ,qi ,ti) selects responses
i is truthful if ti i (i(hi ,qi ,ti))
• truthful strategies must be history independent
(Determ.) strategy profile induces history h• if h is terminal, then quasi-linear utility realized
• if history is unbounded, then assume utility = 0
Regret-based PRM allocation defined as in one-shot
92
Max VCG Payment Scheme
Assume terminal history h• let be revealed PTV at h, x*() be allocation
Max VCG payment scheme:
• where VCG payment is:
93
Incentive Properties
Suppose we elicit type info until MMR allocation has max regret and we use “max VCG”
Define:
Thm: m is -efficient, -ex post IR and
(+(x*()))-ex post IC.
• weaker results due to possible payment manipulation
94
Elicitation Approaches
Two Phases:
Standard max regret based approaches• give us bounds on efficiency , no a priori bounds
Regret-based followed by payment elicitation• once small enough, elicit additional payment
information until max is small enough
95
Elicitation Approaches
Direct optimization:
global manipulability:u(best lie) - u(truth)
ask queries that directly reduce global manipulability
can be formulated as regret-style optimizationanalogous query strategies possible
96
Test Domains
Car Rental Problem:• 1 client , 2 dealers• GAI valuation/costs: 13 factors, size 1-4• Car: 8 attributes, 2-9 values• Total 825 parameters
Small Random Problems:• supplier-selection, 1 buyer, 2 sellers• 81 parameters
97
Results: Car Rental
Initial regret: 99% of opt SWZero-regret: 71/77 queriesAvg remaining uncertainty: 92% vs 64% at zero-manipulability Avg nb params queried: 8%
•relevant parameters
•reduces revelation
•improves decision quality
98
Results: Random Problems
99
ContributionsTheoretical framework for Partial Revelation Mech DesignOne-shot mechanisms
• generalize VCG to PRMs (allocation + payments)• v. general payments: secondary objectives• algorithm to design partial types
Sequential mechanisms• slightly different model, but similar results• algorithm to design query strategy
Viewpoint: why approximate incentives are useful• Approximate decision trade off cost vs. quality
• Formal, approximate IC ensures practical, exact IC
Applicable to general Mechanism Design problemsEmpirically very effective
100
PRMs: Future WorkFurther investigate splitting / elicitation heuristicsMore experimentation
• Larger problems• Combinatorial Auctions
Formal model manipulability cost formal, exact IC
Formal model revelation costs explicit revelation vs efficiency trade-off
Sequentially optimal elicitation
101
Questions ?
References:
Nathanael Hyafil and Craig Boutilier
• “Regret-based Incremental Partial Revelation
Mechanisms”, AAAI 2006
• “One-shot Partial Revelation Mechanisms”,
Working Paper, 2006
102
Extra Slides - Part 1
103
Fishburn [1967]: Default Outcomes
Define default outcome:For any x, let x[I] be restriction of x to vars I, with remaining replaced by default values:
Utility of x can be written [F67]:
• sum of utilities of certain related “key” outcomes
104
Key Outcome Decomposition
Example: GAI over I={ABC}, J={BCD}, K={DE}
u(x) = u(x[I]) + u(x[J]) + u(x[K])
- u(x[IJ]) - u(x[IK]) - u(x[JK])
+ u(x[IJK])
u(abcde) = u(x[abc]) + u(x[bcd]) + u(x[de])
- u(x[bc]) - u(x[]) - u(x[d])
+ u(x[])
u(abcde) = u(abcd0e0) + u(a0bcde0) + u(a0b0c0de)
- u(a0bcd0e0) - u(a0b0c0de0)
105
Canonical Decompostion [F67]
This leads to canonical decompostion of u:
u1(x1, x2) u2(x2, x3)
u(abcde) = u(abcd0e0)
+ u(a0bcde0) - u(a0bcd0e0)
+ u(a0b0c0de) - u(a0b0c0de0)
e.g., I={ABC}, J={BCD}, K={DE}
106
Local Queries
Thus, if for some y (where Y =X \ Xi \ C(Xi) )
then for all y’
hence we can legitimately ask local queries:
107
Implications for Minimax RegretComplicates MMR
• utility of outcome depends linearly on GAI parameters
• but GAI parameters depend on bounds induced by two types of queries: quadratic constraints
Local pairwise regret notion can be modified
• To compute rx[k]x’[k] set values: vx’[k] to u.b. and vx[k] to l.b.
If vx’[k]↑ > vx[k]↓: max u(xT[k]) and min u(x[k])
otherwise do the opposite
108
Bayesian Utility Function Uncertainty
User’s actual utility u unknown
Assume density P over U = [0,1]n
Given belief state P, EU of decision x is:
Decision making is easy, but elicitation harder?• must assess expected value of information in query
U x uPupPxEU )(),(
109
Extra Slides - Part 2
Top Related