Results 1  10
of
37
Choosing Samples to Compute HeuristicStrategy Nash
 In Fifth Workshop on AgentMediated Electronic Commerce
, 2003
"... Auctions define games of incomplete information for which it is often too hard to compute the exact BayesianNash equilibrium. Instead, the infinite strategy space is often populated with heuristic strategies, such as myopic bestresponse to prices. Given these heuristic strategies, it can be usefu ..."
Abstract

Cited by 30 (0 self)
 Add to MetaCart
Auctions define games of incomplete information for which it is often too hard to compute the exact BayesianNash equilibrium. Instead, the infinite strategy space is often populated with heuristic strategies, such as myopic bestresponse to prices. Given these heuristic strategies, it can be useful to evaluate the strategies and the auction design by computing a Nash equilibrium across the restricted strategy space. First, it is necessary to compute the expected payoff for each heuristic strategy profile. This step involves sampling the auction and averaging over multiple simulations, and its cost can dominate the cost of computing the equilibrium given a payoff matrix. In this paper, we propose two information theoretic approaches to determine the next sample through an interleaving of equilibrium calculations and payoff refinement. Initial experiments demonstrate that both methods reduce error in the computed Nash equilibrium as samples are performed at faster rates than naive uniform sampling. The second, faster method, has a lower metadeliberation cost and better scaling properties. We discuss how our sampling methodology could be used within experimental mechanism design.
Selfish Grid Computing: GameTheoretic Modeling and NAS Performance Results
 in Proceedings of the International Symposium on Cluster Computing and the Grid (CCGrid2005
, 2005
"... Selfish behaviors of individual machines in a Grid can potentially damage the performance of the system as a whole. However, scrutinizing the Grid by taking into account the noncooperativeness of machines is a largely unexplored research problem. In this paper, we first present a new hierarchical ga ..."
Abstract

Cited by 16 (6 self)
 Add to MetaCart
Selfish behaviors of individual machines in a Grid can potentially damage the performance of the system as a whole. However, scrutinizing the Grid by taking into account the noncooperativeness of machines is a largely unexplored research problem. In this paper, we first present a new hierarchical gametheoretic model of the Grid that matches well with the physical administrative structure in reallife situations. We then focus on the impact of selfishness in intrasite job execution mechanisms. Based on our novel utility functions, we analytically derive the Nash equilibrium and optimal strategies for the general case.
Selfconfirming price prediction for bidding in simultaneous ascending auctions
 In TwentyFirst Conference on Uncertainty in Artificial Intelligence
, 2005
"... Simultaneous ascending auctions present agents with the exposure problem: bidding to acquire a bundle risks the possibility of obtaining an undesired subset of the goods. Auction theory provides little guidance for dealing with this problem. We present a new family of decisiontheoretic bidding strat ..."
Abstract

Cited by 14 (6 self)
 Add to MetaCart
Simultaneous ascending auctions present agents with the exposure problem: bidding to acquire a bundle risks the possibility of obtaining an undesired subset of the goods. Auction theory provides little guidance for dealing with this problem. We present a new family of decisiontheoretic bidding strategies that use probabilistic predictions of final prices. We focus on selfconfirming price distribution predictions, which by definition turn out to be correct when all agents bid decisiontheoretically based on them. Bidding based on these is provably not optimal in general, but our experimental evidence indicates the strategy can be quite effective compared to
Learning payoff functions in infinite games
 In Nineteenth International Joint Conference on Artificial Intelligence
, 2005
"... We consider a class of games with realvalued strategies and payoff information available only in the form of data from a given sample of strategy profiles. Solving such games with respect to the underlying strategy space requires generalizing from the data to a complete payofffunction representati ..."
Abstract

Cited by 13 (6 self)
 Add to MetaCart
We consider a class of games with realvalued strategies and payoff information available only in the form of data from a given sample of strategy profiles. Solving such games with respect to the underlying strategy space requires generalizing from the data to a complete payofffunction representation. We address payofffunction learning as a standard regression problem, with provision for capturing known structure (symmetry) in the multiagent environment. To measure learning performance, we consider the relative utility of prescribed strategies, rather than the accuracy of payoff functions per se. We demonstrate our approach and evaluate its effectiveness on two examples: a twoplayer version of the firstprice sealedbid auction (with known analytical form), and a fiveplayer marketbased scheduling game (with no known solution). 1
Zip60: Further explorations in the evolutionary design of online auction market mechanisms
, 2005
"... The “ZIP ” adaptive automated trading algorithm has been demonstrated to outperform human traders in experimental studies of continuous double auction (CDA) markets populated by mixtures of human and “software robot ” traders. Previous papers have shown that values of the eight parameters governing ..."
Abstract

Cited by 9 (4 self)
 Add to MetaCart
The “ZIP ” adaptive automated trading algorithm has been demonstrated to outperform human traders in experimental studies of continuous double auction (CDA) markets populated by mixtures of human and “software robot ” traders. Previous papers have shown that values of the eight parameters governing behavior of ZIP traders can be automatically optimized using a genetic algorithm (GA), and that markets populated by GAoptimized traders perform better than those populated by ZIP traders with manuallyset parameter values. This paper introduces a more sophisticated version of the ZIP algorithm, called “ZIP60”, which requires the values of 60 parameters to be set correctly. ZIP60 is shown here to produce significantly better results in comparison to the original ZIP algorithm (called “ZIP8 ” hereafter) when a GA is used to search the 60dimensional parameter space. It is also demonstrated here that this works best when the GA itself has control over the dimensionality of the searchspace, allowing evolution to guide the expansion of the searchspace up from 8 parameters to 60 via intermediate steps. Principal component analysis of the best evolved ZIP60 parametersets establishes that no ZIP8 solutions are embedded in the 60dimensional space. Moreover, some of the results and analysis presented here
Selfish Grids: GameTheoretic Modeling and NAS/PSA Benchmark Evaluation
"... Abstract—Selfish behaviors of individual machines in a Grid can potentially damage the performance of the system as a whole. However, scrutinizing the Grid by taking into account the noncooperativeness of machines is a largely unexplored research problem. In this paper, we first present a new hierar ..."
Abstract

Cited by 8 (1 self)
 Add to MetaCart
Abstract—Selfish behaviors of individual machines in a Grid can potentially damage the performance of the system as a whole. However, scrutinizing the Grid by taking into account the noncooperativeness of machines is a largely unexplored research problem. In this paper, we first present a new hierarchical gametheoretic model of the Grid that matches well with the physical administrative structure in reallife situations. We then focus on the impact of selfishness in intrasite job execution mechanisms. Based on our novel utility functions, we analytically derive the Nash equilibrium and optimal strategies for the general case. To study the effects of different strategies, we have also performed extensive simulations by using a wellknown practical scheduling algorithm over the NAS (Numerical Aerodynamic Simulation) and the PSA (Parameter Sweep Application) workloads. We have studied the overall job execution performance of the Grid system under a wide range of parameters. Specifically, we find that the Optimal selfish strategy significantly outperforms the Nash selfish strategy. Our performance evaluation results can serve as a valuable reference for designing appropriate strategies in a practical Grid.
Using tabu bestresponse search to find pure strategy Nash equilibria in normal form games
 In Fourth International Joint Conference on Autonomous Agents and Multiagent Systems
"... We present a new method for computing pure strategy Nash equilibria for a class of nperson games where it is computationally expensive to compute the payoffs of the players as a result of the joint actions. Previous algorithms to compute Nash equilibria are based on mathematical programming and ana ..."
Abstract

Cited by 8 (0 self)
 Add to MetaCart
We present a new method for computing pure strategy Nash equilibria for a class of nperson games where it is computationally expensive to compute the payoffs of the players as a result of the joint actions. Previous algorithms to compute Nash equilibria are based on mathematical programming and analytical derivation, and require a complete payoff matrix as input. However, determining a payoff matrix can itself be computationally intensive, as is the case with combinatorial auctions. This paper proposes an approach, based on bestresponse dynamics and tabu search, that avoids the requirement that we have a complete payoff matrix upfront, and instead computes the payoffs only as they become relevant to the search. The tabu features help break bestresponse cycles, and allow the algorithm to find pure strategy Nash equilibria in multiplayer games where bestresponse would typically fail. We test the algorithm on several classes of standard and random games, and present empirical results that show the algorithm performs well and gives the designer control over the tradeoffs between search time and completeness.
An OptionsBased Solution to the Sequential Auction Problem
 ARTIFICIAL INTELLIGENCE
, 2009
"... ..."
Using a Market Economy to Provision Compute Resources Across Planetwide Clusters
"... Abstract—We present a practical, marketbased solution to the resource provisioning problem in a set of heterogeneous resource clusters. We focus on provisioning rather than immediate scheduling decisions to allow users to change longterm job specifications based on market feedback. Users enter bid ..."
Abstract

Cited by 7 (1 self)
 Add to MetaCart
Abstract—We present a practical, marketbased solution to the resource provisioning problem in a set of heterogeneous resource clusters. We focus on provisioning rather than immediate scheduling decisions to allow users to change longterm job specifications based on market feedback. Users enter bids to purchase quotas, or bundles of resources for longterm use. These requests are mapped into a simulated clock auction which determines uniform, fair resource prices that balance supply and demand. The reserve prices for resources sold by the operator in this auction are set based on current utilization, thus guiding the users as they set their bids towards underutilized resources. By running these auctions at regular time intervals, prices fluctuate like those in a realworld economy and provide motivation for users to engineer systems that can best take advantage of available resources. These ideas were implemented in an experimental resource market at Google. Our preliminary results demonstrate an efficient transition of users from more congested resource pools to less congested resources. The disparate engineering costs for users to reconfigure their jobs to run on less expensive resource pools was evidenced by the large price premiums some users were willing to pay for more expensive resources. The final resource allocations illustrated how this framework can lead to significant, beneficial changes in user behavior, reducing the excessive shortages and surpluses of more traditional allocation methods. I.