Robust Mechanisms for Information Elicitation Aviv Zohar & Jeffrey S. Rosenschein.

Slides:



Advertisements
Similar presentations
Eliciting Honest Feedback 1.Eliciting Honest Feedback: The Peer-Prediction Model (Miller, Resnick, Zeckhauser) 2.Minimum Payments that Reward Honest Feedback.
Advertisements

Ulams Game and Universal Communications Using Feedback Ofer Shayevitz June 2006.
H.S. Physical Science Chapters 1 and 2
Nash’s Theorem Theorem (Nash, 1951): Every finite game (finite number of players, finite number of pure strategies) has at least one mixed-strategy Nash.
Non myopic strategy Truth or Lie?. Scoring Rules One important feature of market scoring rules is that they are myopic strategy proof. That means that.
Chapter 37 Asymmetric Information In reality, it is often the case that one of the transacting party has less information than the other. Consider a market.
Auctions. Strategic Situation You are bidding for an object in an auction. The object has a value to you of $20. How much should you bid? Depends on auction.
An Approximate Truthful Mechanism for Combinatorial Auctions An Internet Mathematics paper by Aaron Archer, Christos Papadimitriou, Kunal Talwar and Éva.
Least squares CS1114
SA-1 Probabilistic Robotics Planning and Control: Partially Observable Markov Decision Processes.
1 Regret-based Incremental Partial Revelation Mechanism Design Nathanaël Hyafil, Craig Boutilier AAAI 2006 Department of Computer Science University of.
Preference Elicitation Partial-revelation VCG mechanism for Combinatorial Auctions and Eliciting Non-price Preferences in Combinatorial Auctions.
Factoring Polynomials
Game Theory, Mechanism Design, Differential Privacy (and you). Aaron Roth DIMACS Workshop on Differential Privacy October 24.
1. problem set 12 from Binmore’s Fun and Games. p.564 Ex. 41 p.565 Ex. 42.
Satisfaction Equilibrium Stéphane Ross. Canadian AI / 21 Problem In real life multiagent systems :  Agents generally do not know the preferences.
Bundling Equilibrium in Combinatorial Auctions Written by: Presented by: Ron Holzman Rica Gonen Noa Kfir-Dahav Dov Monderer Moshe Tennenholtz.
Adverse Selection Asymmetric information is feature of many markets
Planning under Uncertainty
POMDPs: Partially Observable Markov Decision Processes Advanced AI
458 Interlude (Optimization and other Numerical Methods) Fish 458, Lecture 8.
1 Introduction to Linear and Integer Programming Lecture 9: Feb 14.
Mechanism Design. Overview Incentives in teams (T. Groves (1973)) Algorithmic mechanism design (Nisan and Ronen (2000)) - Shortest Path - Task Scheduling.
Agent Technology for e-Commerce Chapter 10: Mechanism Design Maria Fasli
Computational Methods for Management and Economics Carla Gomes
Extensions to Consumer theory Inter-temporal choice Uncertainty Revealed preferences.
Sequences of Take-It-or-Leave-it Offers: Near-Optimal Auctions Without Full Valuation Revelation Tuomas Sandholm and Andrew Gilpin Carnegie Mellon University.
APEC 8205: Applied Game Theory Fall 2007
Robust Mechanisms for Information Elicitation Aviv Zohar & Jeffrey S. Rosenschein The Hebrew University.
Incomplete Contracts Renegotiation, Communications and Theory December 10, 2007.
Trust Based Mechanism Design. Use MD Motivation Fuse the fields of trust-modelling and mechanism design Trust measures how good an interaction partner.
Physics and Measurements.
Complexity of Mechanism Design Vincent Conitzer and Tuomas Sandholm Carnegie Mellon University Computer Science Department.
Foundations of Privacy Lecture 11 Lecturer: Moni Naor.
Yang Cai Sep 15, An overview of today’s class Myerson’s Lemma (cont’d) Application of Myerson’s Lemma Revelation Principle Intro to Revenue Maximization.
Daniel Kroening and Ofer Strichman Decision Procedures An Algorithmic Point of View Deciding ILPs with Branch & Bound ILP References: ‘Integer Programming’
Game Theory.
The Marriage Problem Finding an Optimal Stopping Procedure.
COMP14112: Artificial Intelligence Fundamentals L ecture 3 - Foundations of Probabilistic Reasoning Lecturer: Xiao-Jun Zeng
Auction Theory Class 2 – Revenue equivalence 1. This class: revenue Revenue in auctions – Connection to order statistics The revelation principle The.
Principal - Agent Games. Sometimes asymmetric information develops after a contract has been signed In this case, signaling and screening do not help,
Decision Procedures An Algorithmic Point of View
The Hat Game 11/19/04 James Fiedler. References Hendrik W. Lenstra, Jr. and Gadiel Seroussi, On Hats and Other Covers, preprint, 2002,
Mechanisms for Making Crowds Truthful Andrew Mao, Sergiy Nesterko.
CPS 173 Mechanism design Vincent Conitzer
The Multiplicative Weights Update Method Based on Arora, Hazan & Kale (2005) Mashor Housh Oded Cats Advanced simulation methods Prof. Rubinstein.
© 2009 Institute of Information Management National Chiao Tung University Lecture Note II-3 Static Games of Incomplete Information Static Bayesian Game.
Sequences of Take-It-or-Leave-it Offers: Near-Optimal Auctions Without Full Valuation Revelation Tuomas Sandholm and Andrew Gilpin Carnegie Mellon University.
Auction Seminar Optimal Mechanism Presentation by: Alon Resler Supervised by: Amos Fiat.
Standard and Extended Form Games A Lesson in Multiagent System Based on Jose Vidal’s book Fundamentals of Multiagent Systems Henry Hexmoor, SIUC.
Section 3.1: Proof Strategy Now that we have a fair amount of experience with proofs, we will start to prove more difficult theorems. Our experience so.
Auction Theory תכנון מכרזים ומכירות פומביות Topic 7 – VCG mechanisms 1.
Moshe Tennenholtz, Aviv Zohar Learning Equilibria in Repeated Congestion Games.
Yang Cai Oct 08, An overview of today’s class Basic LP Formulation for Multiple Bidders Succinct LP: Reduced Form of an Auction The Structure of.
Chapters 29, 30 Game Theory A good time to talk about game theory since we have actually seen some types of equilibria last time. Game theory is concerned.
Regret Minimizing Equilibria of Games with Strict Type Uncertainty Stony Brook Conference on Game Theory Nathanaël Hyafil and Craig Boutilier Department.
Topic 3 Games in Extensive Form 1. A. Perfect Information Games in Extensive Form. 1 RaiseFold Raise (0,0) (-1,1) Raise (1,-1) (-1,1)(2,-2) 2.
Data Analysis Econ 176, Fall Populations When we run an experiment, we are always measuring an outcome, x. We say that an outcome belongs to some.
Automated Mechanism Design Tuomas Sandholm Presented by Dimitri Mostinski November 17, 2004.
Manipulating the Quota in Weighted Voting Games (M. Zuckerman, P. Faliszewski, Y. Bachrach, and E. Elkind) ‏ Presented by: Sen Li Software Technologies.
Decision Trees Binary output – easily extendible to multiple output classes. Takes a set of attributes for a given situation or object and outputs a yes/no.
Part 3 Linear Programming
Advanced Engineering Mathematics, 7 th Edition Peter V. O’Neil © 2012 Cengage Learning Engineering. All Rights Reserved. CHAPTER 4 Series Solutions.
Reconstructing Preferences from Opaque Transactions Avrim Blum Carnegie Mellon University Joint work with Yishay Mansour (Tel-Aviv) and Jamie Morgenstern.
Computation, The Missing Ingredient in Classical Economics Edward Tsang Centre for Computational Finance and Economic Agents (CCFEA) University of Essex.
CPS Mechanism design Michael Albert and Vincent Conitzer
Robust Mechanism Design with Correlated Distributions
Vincent Conitzer Mechanism design Vincent Conitzer
Vincent Conitzer CPS 173 Mechanism design Vincent Conitzer
Reinforcement Learning Dealing with Partial Observability
Presentation transcript:

Robust Mechanisms for Information Elicitation Aviv Zohar & Jeffrey S. Rosenschein

Overview of the talk Motivation – how to pay for information Scoring Rules Mechanisms for information elicitation Robust mechanisms for a single agent Multi-agent extensions Conclusion

Motivation - How to Pay for Information Alice wishes to know the weather in Tel- Aviv –This cannot be predicted in advance by anyone! She’s only interested in the chance of rain. There are two options:

Paying for Information Bob lives in Tel-Aviv. He can go outside and check the weather. Getting the information costs him some effort. A cost of c. He wants Alice to pay him for his efforts.

Paying for Information If Alice pays him c$ no matter what he says he can just make something up. If Alice pays him c 1 $ for saying rain, and c 2 $ for saying clear weather, he will pick the larger payment every time. Conclusion: Alice has to have some way to verify the information. Example: The weather in Jerusalem is correlated with the weather in Tel-Aviv In the real world we often buy unverified information. We are usually playing a repeated game.

How to Pay for Information Bob knows that Tel-Aviv is usually sunny. His beliefs affect the cost-benefit analysis. Alice needs to take Bob’s beliefs into consideration when deciding on payments. Does she know what Bob believes? Usually only approximately! Can she find a payment scheme to Bob that will be robust against small changes in belief?

Information Elicitation vs. Preference Elicitation. Information Elicitation: –Knowledge is changing hands. –The seller only cares about payment. –Not interested in how the knowledge is to be used. –The buyer wants the truth! Preference Elicitation: –Information is just the means to an end: Achieving some optimal outcome. –The outcome is the bottom line. –The mechanism has more freedom of action – can control outcome as well as payments. (EXAMPLE: Auctions)

The Direct Revelation Principle If any mechanism exists for the problem then there is a mechanism in which the participants reveal everything. A1 Mech. A3 A2 Outcome+ Payments a1 Mech. a3 a2 Outcome+ Payments A1A3 A2

Direct Revelation Direct revelation can allow us to get over the problem of learning bob’s beliefs. We can ask Bob to reveal everything: –The information to be sold –His beliefs about probabilities. But… –We don’t want to reveal everything. –Information is what Bob sells! –No trusted third party.

Proper Scoring Rules A way to evaluate a probabilistic prediction. For a prediction p, and a final outcome o we shall pay: S(p,o). A proper scoring rule is one in which telling a more accurate prediction gives a higer payment: E o~p [S(p,o)] > E o~p [S(q,o)]

Scoring Rules E o~p [S(p,o)] > E o~p [S(q,o)] There are lots of functions S(.,.) that fulfill this condition. Example: Logarithmic payments: S(p,i) = log(p i ) In which case:

Scoring Rules Predictions are given in the form of probability distributions. How do we combine the predictions of two different experts that have access to different sources of information? We need a model of how their information interacts.

Our Model Seller i owns a random variable X i that it pays c i to discover. Buyer owns a random variable Ω After it learns about values x’ from the sellers and a value ω, it pays the sellers u i ω,x’ The variables X 1,X 2,…,Ω are presumably not independent We assume that they are governed by probability distribution p x1,x2,…,ω Now we know how to combine information from several sources. Pr(ω|X 1, X 2 …)

The Model X1X1 Seller 1 Buyer Ω X2X2 Seller 2 c1c1 c2c2

The Requirements from a Proper Mechanism (Single Agent) 1.Truth-telling: The truth is more profitable than any lie. 2.Investment: Knowing is better than guessing. 3.Individual Rationality: There is a positive expected gain from participating.

Finding a Mechanism We assume P is known. The constraints are all linear in the payments u. We can find a payment scheme using some LP solver. We can optimize the cost too: When can we find a good mechanism? What is the optimal cost?

The Truth is Enough Suppose we have some set of payments that satisfies the truthfulness constraints: We can scale and shift it To satisfy the other constraints.

The Truthfulness Constraints Let’s define: Now we get: And also: We need every pair of vectors p x,p x’ to be linearly separated by v x,x’

A Geometric Interpretation of Truthfulness V x,x’ p x’ pxpx Notice that there can be many ways to select the separating plane ω1ω1 ω2ω2

Existence of a Mechanism A mechanism exists if and only if all vectors p x are pair-wise independent. –One direction is easy: we can’t separate vectors that are linearly dependent. –For the other direction: show a working mechanism: –Setting does the trick.

Robust Mechanisms We return to the case where P is not known exactly. We assume ε is small (according to L ∞ ). certain solutions may be better than others p x’ pxpx ω1ω1 ω2ω2

Robustness of a Specific Payment Scheme A conservative definition: A payment scheme u will be considered ε-robust with regard to distribution if it is proper for every distribution for which How do we find the robustness level of a payment scheme? –Find the minimal ε for which a constraint is violated.

Robustness of a Payment Scheme The robustness of one of the truthfulness constraints can be found by solving: After solving a similar program for every constraint, take the smallest ε found.

Finding a Robust solution Given an ε, all ε-robust solutions form a convex set. This is a stochastic programming problem. –Find a solution to a mathematical program with uncertainty regarding the constraints. The ellipsoid algorithm needs only a separation oracle in order to optimize over the set of solutions. A separation oracle provides a linear separator between any non-solution and the set of solutions. We’ve just seen how to find one! Find a constraint that the payments + a perturbation violate.

The full stochastic program:

Robust Mechanisms Definition: The robustness level of a problem p is the largest ε that can be set as the robustness of a solution. How can we find it? Use binary search. –The robustness level is somewhere between 0 and 1. –Test at any wanted ε in between by trying to actually find an ε-robust solution. –Then, update the boundaries according to the answer.

A Bound for Problem Robustness Problem robustness is only is only bounded by the truthfulness constraints. –Again, shifting and rescaling takes care of the other constraints. A simple bound can be derived: p x’ pxpx ω2ω2 ε x’ ε xε x

Mechanisms for Multiple Sellers Collusion between agents is a critical matter. If they can move payments and share information, we can treat them as one agent with multiple sources of information. An exponential number of constraints is needed. Tension within the group may limit their collusion. From here on we assume no collusion is possible.

Mechanisms for Multiple Sellers Two main options: 1.Mechanisms that work in only in equilibrium. –Truth telling is profitable only when everyone else does it. –Payments are conditioned on all the information –Other equilibriums may exist. 2.Dominant strategy mechanisms. –It is always better to tell the truth. –Payments are conditioned on the agent’s own information only (And the verifier).

A Simple Example (2x2x2 ) Pr(ω=1|x 1,x 2 )Pr(x 1,x 2 )x2x2 x1x1 01/ δ1/ δ 11 Pr(ω|x2=1)=Pr(ω|x2=0) Hence no dominant strategy mechanism for player 2 But a mechanism in equilibrium exists.

A Simple Example (2x2x2 ) Pr(ω=1|x 1,x 2 )Pr(x 1,x 2 )x2x2 x1x1 01/ δ1/ δ 11 The variable Ω is slightly biased to agree with player 1’s variable. We can have a dominant strategy mechanism for player 1.

Robust Mechanisms for Many Sellers Dominant strategy mechanism – Just like the single agent case. May not always exist. Mechanisms that work in equilibrium- problematic. An equilibrium is a best response to a best response. A player must believe that its counterpart will play the equilibrium strategy. This only happens if it believes that the other believes that it will play the equilibrium. And so on…

Belief Hierarchies Assume player A believes the probability is p player B might conceivably believe it’s p’ Furthermore it may believe that A believes it is p’’. p’’ may be far from p, and we get further away with every step. P’’ P’ P

What can we do? We can consider bounded players. Only look some distance into the belief hierarchy. We can create finite belief hierarchies. –The first player has a dominant strategy. –The payment to second player depends only on the first. –Payment to the third only on the previous two –Etc. Every player considers just beliefs of players that precede him. They do not care about his beliefs. No loops.

Finite belief hierarchies Only a single equilibrium. Very reasonable that it will be played. Such mechanisms might not always exist The extreme case: –All agents have a dominant strategy mechanism.

Conclusion Designing information elicitation mechanisms: –Easy for one agent. –Can be extended easily to robust mechanism –Complicated for many agent. –Robust extension is unclear in equilibrium. –Collusion makes the design even harder.