669 Hits in 6.4 sec

Multi-Agent Learning in Double-side Auctions forPeer-to-peer Energy Trading [article]

Zibo Zhao, Andrew L. Liu
2020 arXiv   pre-print
To address this issue, we propose an automated bidding framework in a repeated auction based on multi-armed bandit learning, which aims to minimize each bidder's cumulative regret.  ...  An alternative to FIT is a market based approach; i.e., consumers and DER owners trade energy in an auction-based peer-to-peer (P2P) market, and the rates are determined by a market clearing process.  ...  prices to bid/ask in a repeated auction setting, we propose a regret-minimizationbased learning approach for a multi-agent system in which bidding/asking prices of agents are automatically chosen by bandit  ... 
arXiv:2002.09435v1 fatcat:g3s4xjgaznbsdbl6nzm64dqbje

Ex-post regret heuristics under private values (I): Fixed and random matching

Rene Saran, Roberto Serrano
2014 Journal of Mathematical Economics  
The results are illustrated in several applications, including second-price auctions, first-price auctions and Bertrand duopolies.  ...  In contexts in which players have no priors, we analyze a heuristic process based on ex-post regret as a guide to understand how to play games of incomplete information under private values.  ...  However, in general games, minimal CUSOAEBR sets can be rather large. We illustrate this by exhibiting the unique minimal CU-SOAEBR sets for the first-price auction and Bertrand duopoly game.  ... 
doi:10.1016/j.jmateco.2014.06.006 fatcat:44jy7k5zafdg5nk6lvdgdata4e

Risk- & Regret-Averse Bidders in Sealed-Bid Auctions

Takashi Hayashiy, Hisayuki Yoshimoto
2014 Social Science Research Network  
Such experimental phenomena is called overbidding, and observed in first-price auction experiments conducted by Cox, Robertson, and Smith (1982) [9] and Cox, Simth, and Walker (1988) [10].  ...  To solve this discrepancy, we construct a structural model of bidding behavior in sealed-bid auctions, one in which bidders may regret their decisions.  ...  We can define wnning and loser regret in the following manners. 58 This subsection can be skepped if a reader is not interested in a third-price auction. 59 More generally, a k-th price auction where k  ... 
doi:10.2139/ssrn.2400092 fatcat:adfnmgxjmvhdzcdvxj2idk47yi

Online Learning of Optimal Bidding Strategy in Repeated Multi-Commodity Auctions [article]

Sevi Baltaoglu, Lang Tong, Qing Zhao
2017 arXiv   pre-print
By showing that the regret is lower bounded by Ω(√(T)) for any strategy, we conclude that DPDS is order optimal up to a √(T) term.  ...  We study the online learning problem of a bidder who participates in repeated auctions.  ...  This work was supported in part by the National Science Foundation under Award 1549989 and by the Army Research Laboratory Network Science CTA under Cooperative Agreement W911NF-09-2-0053.  ... 
arXiv:1703.02567v5 fatcat:jb7wwbg2ezhcjkvzj4dujch3j4

Strategic analysis with simulation-based games

Yevgeniy Vorobeychik, Michael P. Wellman
2009 Proceedings of the 2009 Winter Simulation Conference (WSC)  
We present an overview of an emerging methodology for applying game-theoretic analysis to strategic environments described by a simulator.  ...  We first introduce the problem of solving a simulation-based game, and proceed to review convergence results and confidence bounds about game-theoretic equilibrium estimates.  ...  Since their beginnings these auctions have undergone a series of changes, from a first-price to a next-price format, from rank-by-bid to rank-by-revenue.  ... 
doi:10.1109/wsc.2009.5429347 dblp:conf/wsc/VorobeychikW09 fatcat:nrrgdxyymjghbbl5hqubqp4igu

Budget Pacing in Repeated Auctions: Regret and Efficiency without Convergence [article]

Jason Gaitonde, Yingkai Li, Bar Light, Brendan Lucier, Aleksandrs Slivkins
2022 arXiv   pre-print
This result is robust to the correlation structure between agent valuations and holds for any core auction, a broad class of auctions that includes first-price, second-price, and generalized second-price  ...  This generalizes known regret guarantees for bidders facing stochastic bidding environments in two ways: it applies to a wider class of auctions than previously known, and it allows the environment to  ...  JG is supported in part by NSF Award CCF-1408673 and AFOSR Award FA9550-19-1-0183.  ... 
arXiv:2205.08674v1 fatcat:vvyynebzrnewlotfjefzxitkja

Learning in Games: Robustness of Fast Convergence [article]

Dylan J. Foster, Zhiyuan Li, Thodoris Lykouris, Karthik Sridharan, Eva Tardos
2016 arXiv   pre-print
We show that learning algorithms satisfying a low approximate regret property experience fast convergence to approximate optimality in a large class of repeated games.  ...  Our property, which simply requires that each learner has small regret compared to a (1+ϵ)-multiplicative approximation to the best action in hindsight, is ubiquitous among learning algorithms; it is satisfied  ...  Acknowledgements We thank Vasilis Syrgkanis for sharing simulation software and the NIPS reviewers for pointing out that the GREEN algorithm [AAGO06] satisfies the Low Approximate Regret property.  ... 
arXiv:1606.06244v4 fatcat:h2bvroidizgfni7rclrpa7b66y

Wireless Edge-Empowered Metaverse: A Learning-Based Incentive Mechanism for Virtual Reality [article]

Minrui Xu, Dusit Niyato, Jiawen Kang, Zehui Xiong, Chunyan Miao, Dong In Kim
2021 arXiv   pre-print
., sellers), we design a double Dutch auction mechanism to determine optimal pricing and allocation rules in this market.  ...  by different providers.  ...  The buy-bid of VR users m at time slot t, which is denoted by k t m , represents the maximum price that VR user m is willing to pay for the VR service from SPs, which can be calculated as k t m = v b m  ... 
arXiv:2111.03776v1 fatcat:6vr4ydp35jctznjxila2l5yzty

Real-Time Optimisation for Online Learning in Auctions [article]

Lorenzo Croissant, Marc Abeille, Clément Calauzènes
2020 arXiv   pre-print
In display advertising, a small group of sellers and bidders face each other in up to 10 12 auctions a day.  ...  In this context, revenue maximisation via monopoly price learning is a high-value problem for sellers. By nature, these auctions are online and produce a very high frequency stream of data.  ...  As for other algorithms, by using a doubling trick to handle an unknown horizon, ETC+V-CONV-OGA exhibits a sub-linear regret.  ... 
arXiv:2010.10070v1 fatcat:5raojdsypvdozo4pbcjc7ya5eq

Ex-Post Regret Learning in Games with Fixed and Random Matching: The Case of Private Values

Rene Saran, Roberto Serrano
2010 Social Science Research Network  
The results are illustrated in second-price auctions, first-price auctions and Bertrand duopolies.  ...  In contexts in which players have no priors, we analyze a learning process based on ex-post regret as a guide to understand how to play games of incomplete information under private values.  ...  Examples will be provided in our subsections on applications. 3.1.1 Applications: Second-Price Auctions, First-Price Auctions and Bertrand Duopolies We now illustrate our results by studying specific  ... 
doi:10.2139/ssrn.1622242 fatcat:mk6bfrcgmfatrplkru5tk62bla

Differentially Private Call Auctions and Market Impact [article]

Emily Diana, Hadi Elzayn, Michael Kearns, Aaron Roth, Saeed Sharifi-Malvajerdi, Juba Ziani
2020 arXiv   pre-print
We propose and analyze differentially private (DP) mechanisms for call auctions as an alternative to the complex and ad-hoc privacy efforts that are common in modern electronic markets.  ...  We analyze the incentive properties of our mechanisms and their behavior under natural no-regret learning dynamics by market participants.  ...  Finally, while [Chen et al., 2018] have shown how to privately compute near-optimal prices in double auctions, their process does not guarantee end-to-end joint differentially privacy when taking trade  ... 
arXiv:2002.05699v1 fatcat:yyqhl2ykezd6hjazygybqnfjke

On the Robustness of Second-Price Auctions in Prior-Independent Mechanism Design [article]

Jerry Anunrojwong, Santiago Balseiro, Omar Besbes
2022 arXiv   pre-print
Furthermore, we show that the minimax optimal mechanisms have a simple form across all settings: a second-price auction with random reserve prices, which shows its robustness in prior-independent mechanism  ...  In particular, we show that the first three classes admit the same minimax regret value, which is decreasing with the number of competitors, while the last two have the same minimax regret equal to that  ...  In particular, a reserve price is very valuable in prior-independent environments and the use of a randomized reserve price is also critical to minimize the worst-case regret.  ... 
arXiv:2204.10478v1 fatcat:xso6ogrjsnf4plaw4kkdnrpwoe

Learning Prices for Repeated Auctions with Strategic Buyers [article]

Kareem Amin, Afshin Rostamizadeh, Umar Syed
2013 arXiv   pre-print
a posted-price mechanism.  ...  We model the buyer as a strategic agent, whose goal is to maximize her long-term surplus, and we are interested in mechanisms that maximize the seller's long-term revenue.  ...  Acknowledgements We thank Corinna Cortes, Gagan Goel, Yishay Mansour, Hamid Nazerzadeh and Noam Nisan for early comments on this work and pointers to relevent literature.  ... 
arXiv:1311.6838v1 fatcat:dxfmprptpfdfpkfh4vxblpzebq

Learning to Bid Without Knowing your Value [article]

Zhe Feng, Chara Podimata, Vasilis Syrgkanis
2018 arXiv   pre-print
hindsight, that are exponentially faster in convergence in terms of dependence on the action space, than what would have been derived by applying a generic bandit algorithm and almost equivalent to what  ...  We leverage the structure of the utility of the bidder and the partial feedback that bidders typically receive in auctions, in order to provide algorithms with regret rates against the best fixed bid in  ...  In Appendix E, we also show how to deal with unknown parameters L, T and ∆ o by applying a standard doubling trick. Example 5.5 (First Price and All-Pay Auctions).  ... 
arXiv:1711.01333v5 fatcat:lexaxaulhfbn3e27fm2gnbjjbi

An Efficient and Strategy-Proof Double-Track Auction for Substitutes and Complements

Ning Sun, Zaifu Yang
2014 Social Science Research Network  
In each round of the auction, the auctioneer announces the current prices for all items, bidders respond by reporting their demands at these prices, and then the auctioneer adjusts simultaneously the prices  ...  We prove that although bidders are not assumed to be price-takers and thus can strategically exercise their market power, this dynamic auction always induces the bidders to bid truthfully as price-takers  ...  According to the rule of the double-track auction to be discussed soon in detail, the seller treats A as balanced and so she adjusts p(2) to p(3) = (2, 3) by decreasing the price of B by 1 and holding  ... 
doi:10.2139/ssrn.2492096 fatcat:7xqechc5ozcsvnq22nvdn64zi4
« Previous Showing results 1 — 15 out of 669 results