Distributed Welfare Games

Marden, Jason R.; Wierman, Adam

doi:10.1287/opre.1120.1137

Cited by 175 publications

(172 citation statements)

References 51 publications

Supporting

Mentioning

169

Contrasting

Order By: Relevance

“…This result is an application of the results of [16], as shown for marginal contribution potential games by [5].…”

Section: Theorem 21 In Marginal Contribution Potential Games With Nsupporting

confidence: 53%

“…Furthermore, using marginal contributions to a set function allows us to apply a bound to the game's price of anarchy, which is the ratio of the worst Nash equilibrium to the global optimum [5]. In order to do this, however, we must assume some additional conditions on the problem.…”

Section: Potential Games and Marginal Contribution Utilitiesmentioning

confidence: 99%

“…5,6 In our implementation, the duration of a simulated run is partitioned into two portions: 7 (i) During the first portion, the agents only randomly and independently sample, in order to learn something of their rewards for joint actions (i.e. ε is set to 1 and the Max-Sum algorithm is not used); (ii) In the second portion, at each time step, the Max-Sum algorithm is used to find the greedy action for the ε-greedy learning policy (line 6 of Algorithm 1), while ε values themselves decrease according ε(t) = 1 /8 t −1/J i for all i ∈ N (line 8), which is the schedule used by JSFP, etc.…”

Section: Benchmark Algorithm: Max-sum Q-learningmentioning

confidence: 99%

“…The agents' employ Q -learning to update their reward estimates throughout the entirety of a run. This partition is done 5 The key conceptual difference in the operation of the game-theoretic algorithms discussed in this paper and Max-Sum is that Max-Sum directly computes the optimal value and the argmax variable configuration, whereas the game-theoretic algorithms use the gradient of rewards (i.e. marginal rewards) to find a local optimum.…”

Section: Benchmark Algorithm: Max-sum Q-learningmentioning

confidence: 99%

“…each agent has a decreasing marginal contribution), then the ratio of the worst-case Nash equilibrium to the optimum can be bounded. This ratio is known as the price of anarchy, and [5] show that in submodular marginal contribution games it is at most 1 /2, or in other words, the value of the worst Nash equilibrium solution (local maximum) is within 1 /2 of that of the global optimum. This is effectively a bound on the cost of distributing the control of the problem among multiple autonomous agents.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Learning in Unknown Reward Games: Application to Sensor Networks

Chapman

Leslie²,

Rogers

et al. 2013

The Computer Journal

View full text Add to dashboard Cite

This paper demonstrates a decentralised method for optimisation using game-theoretic multi-agent techniques, applied to a sensor network management problem. Our first major contribution is to show how the marginal contribution utility design is used to construct a unknown-reward potential game formulation of the problem. This formulation exploits the sparse structure of sensor network problems, and allows us to apply a bound to the price of anarchy of the Nash equilibria of the induced game. Furthermore, since the game is a potential game, solutions can be found using multiagent learning techniques. The techniques we derive use Q-learning to estimate an agent's rewards, while an action adaptation process responds to an agent's opponents' behaviour. However, there are many different algorithmic configurations that could be used to solve these games. Thus, our second major contribution is an extensive evaluation of several action adaptation processes. Specifically, we compare six algorithms across a variety of parameter settings to ascertain the quality of the solutions they produce, their speed of convergence, and their robustness to pre-specified parameter choices. Our results show that they each perform similarly across a wide range of parameters. There is, however, a significant effect from moving to a learning policy with sampling probabilities that go to zero too quickly for rewards to be accurately estimated.

show abstract

“…This result is an application of the results of [16], as shown for marginal contribution potential games by [5].…”

Section: Theorem 21 In Marginal Contribution Potential Games With Nsupporting

confidence: 53%

Section: Potential Games and Marginal Contribution Utilitiesmentioning

confidence: 99%

Section: Benchmark Algorithm: Max-sum Q-learningmentioning

confidence: 99%

Section: Benchmark Algorithm: Max-sum Q-learningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Learning in Unknown Reward Games: Application to Sensor Networks

Chapman

Leslie²,

Rogers

et al. 2013

The Computer Journal

View full text Add to dashboard Cite

show abstract

Matrix expression of Owen values

Zhu

Feng

Sun

2021

Asian Journal of Control

View full text Add to dashboard Cite

The Owen value is investigated in this paper. It is a generation of Shapley value as a solution of cooperative games with coalition structures. First, a characteristic function of a cooperative game is expressed as a pseudo‐logical function by using semi‐tensor product of matrices. Then, a matrix formula is given for calculating Owen values. The matrix formula not only makes computing more convenient but also facilitates theoretical analysis. Therefore, an application of Owen value in distributed welfare games is presented. Utility functions of a distributed welfare game with coalition structure are designed based on Owen value. Examples are showed to illustrate the theoretical results.

show abstract

Cooperative caching for content dissemination in vehicular networks

Bitaghsir

Khonsari

2018

Int J Communication

View full text Add to dashboard Cite

A vehicular communication network (VCN) is composed of low-cost base stations called roadside units (RSUs), 3 where each RSU has limited storage space and are able to wirelessly communicate with fast-moving vehicles. Cellular traffic off-loading to vehicular communication networks benefits from the fact that most of the outdoor traffic originates from portable communication devices within vehicles. Further, modern vehicles may be equipped with processing power and storage space better than that of smartphones or tablets, which, in turn, can facilitate advanced signal processing techniques to mitigate channel impairments and achieve higher throughput. Furthermore, vehicular communication technologies are designed for outdoor environments where high Doppler channel conditions are prevalent, which make them preferable over Wi-Fi networks. In this paper, we investigate a content delivery network, in which cellular traffic is being off-loaded via VCNs.We consider a content delivery network where the central content server is hosted at the base station(s) of a cellular network, and the cache servers are hosted at the RSUs of a VCN. A delay tolerant approach to content delivery is explored with a goal of balancing the end-user service quality against the congestion in the cellular network. Upon the request for a file at the central content server from an end-user application, the central content server returns a search interval. The search interval may be chosen as a function of the delay tolerance of the requesting application. During the search interval, the application requesting the file attempts to download the file from the cache servers that the end-user (vehicle) visits as it travels. If the download is not successful by the end of search interval, then the file is delivered over the cellular network from the main server. Since the majority of file downloads are delay tolerant, this approach can considerably reduce the load of the base station.Given the above content delivery model, we want to minimize the cellular traffic, which in turn means to maximize the average hit ratio, the average fraction of total requests served via the local caches per unit time. Given the popularity of the contents, the traffic patterns of streets, and the location of the cache servers (RSUs) in a geographical area, how shall the files be placed in the local cache servers so as to maximize the average hit ratio? We refer to the above problem by cache content placement (CCP) problem. In this paper, we seek to solve CCP in a distributed fashion: design local algorithms that enable each cache server to autonomously choose its files in response to its own local information such that the average hit ratio is maximized. There are several reasons why we prefer distributed approaches over centralized approaches for solving CCP. First, there are several time-varying parameters in vehicular networks, such as file demands or vehicles mobility model, and distributed approaches adapt faster in these dynamic situations compared to the centralized...

show abstract

Distributed Welfare Games

Cited by 175 publications

References 51 publications

Learning in Unknown Reward Games: Application to Sensor Networks

Learning in Unknown Reward Games: Application to Sensor Networks

Matrix expression of Owen values

Cooperative caching for content dissemination in vehicular networks

Contact Info

Product

Resources

About