13,198 Hits in 1.7 sec

Robust Learning of Optimal Auctions [article]

Wenshuo Guo, Michael I. Jordan, Manolis Zampetakis
2021 arXiv   pre-print
We study the problem of learning revenue-optimal multi-bidder auctions from samples when the samples of bidders' valuations can be adversarially corrupted or drawn from distributions that are adversarially  ...  Lastly, we derive sample complexity upper bounds for learning a near-optimal auction for both MHR and regular distributions.  ...  Acknowledgments This work was supported in part by the Mathematical Data Science program of the Office of Naval Research under grant number N00014-18-1-2764.  ... 
arXiv:2107.06259v1 fatcat:w6moynpi2vgbhfgi75et4ehvwe

Adversarial learning for revenue-maximizing auctions [article]

Thomas Nedelec, Jules Baudet, Vianney Perchet, Noureddine El Karoui
2021 arXiv   pre-print
We introduce a new numerical framework to learn optimal bidding strategies in repeated auctions when the seller uses past bids to optimize her mechanism.  ...  We recover essentially all state-of-the-art analytical results for the single-item framework derived previously in the setup where the bidder knows the optimization mechanism used by the seller and extend  ...  Definition 5 ( adversarially-robust learning algorithm).  ... 
arXiv:1909.06806v3 fatcat:wqbfhpmphjctbeptkzas7ehya4

Game Theory Meets Computational Learning Theory (Dagstuhl Seminar 17251)

Paul W. Goldberg, Yishay Mansour, Paul Dütting, Marc Herbstritt
2017 Dagstuhl Reports  
This report documents the program and the outcomes of Dagstuhl Seminar 17251 "Game Theory Meets Computational Learning Theory".  ...  While there have been many Dagstuhl seminars on various aspects of Algorithmic Game Theory, this was the first one to focus on the emerging field of its intersection with computational learning theory.  ...  We investigate this question via two models for studying robust estimation, learning, and optimization.  ... 
doi:10.4230/dagrep.7.6.68 dblp:journals/dagstuhl-reports/GoldbergMD17 fatcat:ca4mfrf3qbdbhhbo7rvc53myti

Dynamic Incentive-Aware Learning: Robust Pricing in Contextual Auctions

Negin Golrezaei, Adel Javanmard, Vahab Mirrokni
2018 Social Science Research Network  
Motivated by pricing in ad exchange markets, we consider the problem of robust learning of reserve prices against strategic buyers in repeated contextual secondprice auctions.  ...  We propose two learning policies that are robust to such strategic behavior.  ...  Thus, while a single shot second-price auction is a truthful mechanism, repeated second-price auctions in which the seller aims at dynamically learning optimal reserve prices of strategic and utility-maximizing  ... 
doi:10.2139/ssrn.3144034 fatcat:qse434acdvdjjorreibgk3t4ri

Deep Reinforcement Learning for Sponsored Search Real-time Bidding [article]

Jun Zhao, Guang Qiu, Ziyu Guan, Wei Zhao, Xiaofei He
2018 arXiv   pre-print
Motivated by the observation that auction sequences of two days share similar transition patterns at a proper aggregation level, we formulate a robust MDP model at hour-aggregation level of the auction  ...  Bidding optimization is one of the most critical problems in online advertising.  ...  In other words, the robust MDP aims to learn the optimal parameter policy to control the real-time bidding model.  ... 
arXiv:1803.00259v1 fatcat:kqjd46oojfdh5p2kin42w7muua

Scanning the Literature

2018 IEEE wireless communications  
To address the disadvantages faced in the UD auction, the authors propose a learning-based UD (LBUD) auction.  ...  Specifically, RDSH simultaneously learns discrete binary codes as well as robust hash functions within a unified model.  ... 
doi:10.1109/mwc.2018.8403921 fatcat:xsweziqrane7lgtlkz7wq77qv4

Certifying Strategyproof Auction Networks [article]

Michael J. Curry, Ping-Yeh Chiang, Tom Goldstein, John Dickerson
2020 arXiv   pre-print
A recent thread of work in "differentiable economics" has used tools from modern deep learning to instead learn good mechanisms.  ...  Optimal auctions maximize a seller's expected revenue subject to individual rationality and strategyproofness for the buyers.  ...  The difficulty of designing optimal auctions has motivated attempts to formulate the auction design problem as a learning problem. Duetting et al.  ... 
arXiv:2006.08742v1 fatcat:2rqer4vtbrdffmxqjhesocl33a

Robust Capacity Control in Revenue Management: A Literature Review

Zheyu Jiang
2018 Open Journal of Business and Management  
In recent 10 years, robust optimization methodology motivated a rapid growing amount of literature on robust revenue management.  ...  structural properties of optimal solution.  ...  These assumptions may not be satisfied in reality, thus study on robust auction mechanism is of great importance.  ... 
doi:10.4236/ojbm.2018.62037 fatcat:bx4k4n2oj5egpcopofgb7yoidy

Multi-Item Mechanisms without Item-Independence: Learnability via Robustness [article]

Johaness Brustle, Yang Cai, Constantinos Daskalakis
2020 arXiv   pre-print
We study the sample complexity of learning revenue-optimal multi-item auctions. We obtain the first set of positive results that go beyond the standard but unrealistic setting of item-independence.  ...  We establish parametrized sample complexity bounds for learning an up-to-ε optimal mechanism in both models, which scale polynomially in the size of the model, i.e. the number of items and bidders, and  ...  Applications of Multi-Item Robustness Lipschitz Continuity of the Optimal Revenue in Multi-item Auctions.  ... 
arXiv:1911.02146v2 fatcat:6y2ojwm27vdc5i4f4jwiwl7a44

Robust Clearing Price Mechanisms for Reserve Price Optimization [article]

Zhe Feng, Sébastien Lahaie
2021 arXiv   pre-print
We propose two simple and computationally efficient methods to set reserve prices based on the notion of a clearing price and make them robust to bidder misreports.  ...  Setting an effective reserve price for strategic bidders in repeated auctions is a central question in online advertising.  ...  ., 2016) show how to learn reserve prices in lazy second price auctions, while (Duetting et al., 2019) learn optimal multi-item auctions using deep learning, subject to incentive compatibility constraints  ... 
arXiv:2107.04638v1 fatcat:lifhihkpc5hijeh3anjnigc6jq

Bidding strategies optimization for the online video ad spot market

Craig Jacobik, Au Dang, Tomu Fujii, Joseph Koerwer, David Schultz, Eric Trouton, William T. Scherer
2011 2011 IEEE Systems and Information Engineering Design Symposium  
The problem is to develop, test, and simulate the best bidding strategies for both second price and first price auctions and then to identify the tradeoffs of using the optimized strategies over current  ...  Tidal TV will then incorporate learning from this exercise to further develop strategies to bid for online video ads.  ...  As a standalone theory, directional learning is not sufficient to produce a robust algorithm. However, directional learning can be combined with impulse balancing to produce an acceptable algorithm.  ... 
doi:10.1109/sieds.2011.5876864 fatcat:lbmkfiwcxnd5zpu6evkwfu5gnq


Meng-Hiot Lim, Steven Gustafson, Natalio Krasnogor, Yew-Soon Ong
2013 Memetic Computing  
Reynolds and Kinnaird-Heether explore the use of auction mechanisms to solve optimization problems.  ...  that are more efficient and robust.  ...  Reynolds and Kinnaird-Heether explore the use of auction mechanisms to solve optimization problems.  ... 
doi:10.1007/s12293-013-0118-2 fatcat:i6uuxqext5g37i543ivrk4drjy

Designing adaptive trading agents

David Pardoe, Peter Stone
2011 ACM SIGecom Exchanges  
The thesis considers how adaptive trading agents can take advantage of previous experience (real or simulated) in other markets while remaining robust in the face of novel situations in a new market.  ...  Its contributions are at the intersection of machine learning and electronic commerce, with particular focus on transfer learning and fully autonomous trading agents.  ...  identify optimal auction parameters when bidder behavior is similar to previous experience, yet the agent remains robust to unexpected situations.  ... 
doi:10.1145/1998549.1998557 fatcat:pjrrjsvawnb3hk2limhyirjxaa

Towards Prior-Free Approximately Truthful One-Shot Auction Learning via Differential Privacy [article]

Daniel Reusche, Nicolás Della Penna
2021 arXiv   pre-print
Designing truthful, revenue maximizing auctions is a core problem of auction design. Multi-item settings have long been elusive.  ...  One remaining problem is to obtain priors in a way that excludes the possibility of manipulating the resulting auctions.  ...  Acknowledgements Daniel Reusche was funded through the NGI0 PET Fund, a fund established by NLnet with financial support from the European Commission's Next Generation Internet programme, under the aegis of  ... 
arXiv:2104.00159v1 fatcat:ihxxgdxryna7dp5ypbi7jahjyy

An Agent Model for First Price and Second Price Private Value Auctions [chapter]

A. J. Bagnall, I. Toft
2004 Lecture Notes in Computer Science  
This paper describes the the private value model of auctions commonly used in auction theory and experimentation and the initial reinforcement learning architecture of the adaptive agent competing in auctions  ...  the process of learning from experience.  ...  Future Work We will continue to experiment with the PVM in order to find the most robust learning mechanism for scenarios with a known optimal strategy.  ... 
doi:10.1007/978-3-540-24621-3_23 fatcat:jrbcvksh2jfbjnpml6z4gdai4q
« Previous Showing results 1 — 15 out of 13,198 results