AI helps you reading Science

AI generates interpretation videos

AI extracts and analyses the key points of the paper to generate videos automatically


pub
Go Generating

AI Traceability

AI parses the academic lineage of this thesis


Master Reading Tree
Generate MRT

AI Insight

AI extracts a summary of this paper


Weibo:
Perhaps autonomous bidding agents will be able to a ect bidding strategies in such future

ATTac-2000: an adaptive autonomous bidding agent

Journal of Artificial Intelligence Research, no. 1 (2001): 238-245

Cited by: 130|Views15
EI

Abstract

The First Trading Agent Competition (TAC) was held from June 22 to July 8, 2000. TAC was designed to create a benchmark problem in the complex domain of e-marketplaces and to motivate researchers to apply unique approaches to a common task. This paper describes \attac, the first-place finisher in TAC. \attac\ uses a principled bidding str...More

Code:

Data:

0
Introduction
  • The rst Trading Agent Competition TAC was held from June 22nd to July 8th, 2000, organized by a group of researchers and developers led by Michael Wellman of the University of Michigan and Peter Wurman of North Carolina State University Wellman, Wurman, O'Malley, Bangera, Lin, Reeves, & Walsh, 2001.
  • A successful agent needed to be able to perform well in any of these possible circumstances
Highlights
  • The rst Trading Agent Competition TAC was held from June 22nd to July 8th, 2000, organized by a group of researchers and developers led by Michael Wellman of the University of Michigan and Peter Wurman of North Carolina State University Wellman, Wurman, O'Malley, Bangera, Lin, Reeves, & Walsh, 2001
  • Their goals included providing a benchmark problem in the complex and rapidly advancing domain of e-marketplaces Eisenberg, 2000 and motivating researchers to apply unique approaches to a common task
  • The top 12 nishers were invited to the semi- nals and nals in Boston, MA on July 8th
  • It is straightforward to write bidding agents to participate in on-line auctions for a single good if the value to the client is xed ahead of time: the agent can bid slightly over the ask price until the auction closes or the price exceeds the value
  • Perhaps autonomous bidding agents will be able to a ect bidding strategies in such future
Results
  • TAC consisted of a preliminary round that ran over the course of a week and involved roughly 80 games for each of the 22 participants.
  • The top 12 nishers were invited to the semi- nals and nals in Boston, MA on July 8th.
  • Since agents and conditions were constantly changing, and since only 13 games were played by each agent in the semi- nals and nals, the competition does not provide a controlled testing environment.
  • ATTac-2000's scores in the 88 preliminary-round games ranged from ,3000 to over 4500 mean 2700, std.
  • A good score in a game instance is in the 3000 to 4000 range.
  • The authors noticed that there were many very bad scores 12 less than 1000 and seven less than 0
Conclusion
  • Conclusion and Future Work

    TAC-2000 was the rst autonomous bidding agent competition. While it was a very successful event, some minor improvements would increase its interest from a multiagent learning perspective.

    Currently, there is no incentive to buy airline tickets until the end of the game.
  • Were there to be information available regarding the bidding behavior of the agents during the game such that other agents could infer clients' preferences, and market supply, demand, and prices, TAC agents would potentially be able to learn to predict market behavior as a game proceeds
  • With or without these modi cations, the authors hope to be able to participate in future TACs, with the goal of adding additional adaptive elements to ATTac-2000.
  • Perhaps autonomous bidding agents will be able to a ect bidding strategies in such future
Summary
  • Introduction:

    The rst Trading Agent Competition TAC was held from June 22nd to July 8th, 2000, organized by a group of researchers and developers led by Michael Wellman of the University of Michigan and Peter Wurman of North Carolina State University Wellman, Wurman, O'Malley, Bangera, Lin, Reeves, & Walsh, 2001.
  • A successful agent needed to be able to perform well in any of these possible circumstances
  • Results:

    TAC consisted of a preliminary round that ran over the course of a week and involved roughly 80 games for each of the 22 participants.
  • The top 12 nishers were invited to the semi- nals and nals in Boston, MA on July 8th.
  • Since agents and conditions were constantly changing, and since only 13 games were played by each agent in the semi- nals and nals, the competition does not provide a controlled testing environment.
  • ATTac-2000's scores in the 88 preliminary-round games ranged from ,3000 to over 4500 mean 2700, std.
  • A good score in a game instance is in the 3000 to 4000 range.
  • The authors noticed that there were many very bad scores 12 less than 1000 and seven less than 0
  • Conclusion:

    Conclusion and Future Work

    TAC-2000 was the rst autonomous bidding agent competition. While it was a very successful event, some minor improvements would increase its interest from a multiagent learning perspective.

    Currently, there is no incentive to buy airline tickets until the end of the game.
  • Were there to be information available regarding the bidding behavior of the agents during the game such that other agents could infer clients' preferences, and market supply, demand, and prices, TAC agents would potentially be able to learn to predict market behavior as a game proceeds
  • With or without these modi cations, the authors hope to be able to participate in future TACs, with the goal of adding additional adaptive elements to ATTac-2000.
  • Perhaps autonomous bidding agents will be able to a ect bidding strategies in such future
Tables
  • Table1: ATTac-2000's client preferences from game 3070. BEV, SEV, and TEV are EVs for baseball, symphony, and theater respectively
  • Table2: ATTac-2000's client allocations and utilities from game 3070. Client 1's B4" under Ent'ment" indicates baseball on day 4
  • Table3: An overview of ATTac-2000's high-level strategy
  • Table4: The scores of the 8 TAC nalists in the semi- nals and nals 13 games
  • Table5: The di erence between ATTac-2000's score and the score of each of the other seven agents averaged over all games in a controlled experiment. All di erences are statistically signi cant at the 0:001 level, except the one marked in italics
Download tables as Excel
Related work
Reference
  • Cramton, P. C. 1997. The FCC spectrum auctions: An early assessment. Journal of Economics and Management Strategy, 63, 431 495.
    Google ScholarLocate open access versionFindings
  • Csirik, J. A., Littman, M. L., Singh, S., & Stone, P. 2001. FAucS: An FCC spectrum auction simulator for autonomous bidding agents. In Proceedings of the Second International Workshop on Electronic Commerce. To appear. Available at. http:www.research.att.com ~pstone papers.html
    Locate open access versionFindings
  • Eisenberg, A. 2000. In online auctions of the future, it'll be bot vs. bot vs. bot. The New York Times. August 17th.
    Google ScholarFindings
  • Garey, M. R., & Johnson, D. S. 1979. Computers and Intractability: A Guide to the Theory of NP-completeness. Freeman, San Francisco, CA.
    Google ScholarFindings
  • Gimenez-Funes, E., Godo, L., Rodriguez-Aguiolar, J. A., & Garcia-Calves, P. 1998. Designing bidding strategies for trading agents in electronic auctions. In Proceedings of the Third International Conference on Multi-Agent Systems, pp. 136 143.
    Google ScholarLocate open access versionFindings
  • Greenwald, A., Boyan, J., Kirby, R. M., & Reiter, J. 2001. Bidding algorithms for simultaneous auctions. In Proceedings of Third ACM Conference on E-Commerce, p. to appear.
    Google ScholarLocate open access versionFindings
  • Greenwald, A., & Kephart, J. O. 1999. Shopbots and pricebots. In Proceedings of the Sixteenth International Joint Conference on Arti cial Intelligence, pp. 506 511.
    Google ScholarLocate open access versionFindings
  • Greenwald, A., & Stone, P. 2001. Autonomous bidding agents in the trading agent competition. IEEE Internet Computing, 52, 52 60.
    Google ScholarLocate open access versionFindings
  • Ito, T., Fukuta, N., Shintani, T., & Sycara, K. 2000. Biddingbot: a multiagent support system for cooperative bidding in multiple auctions. In Proceedings of the Fourth International Conference on MultiAgent Systems, pp. 399 400.
    Google ScholarLocate open access versionFindings
  • Klemperer, P. 1999. Auction theory: A guide to the literature. Journal of Economic Surveys, 133, 227 86.
    Google ScholarLocate open access versionFindings
  • Littman, M. L. 1994. Markov games as a framework for multi-agent reinforcement learning. In Proceedings of the Eleventh International Conference on Machine Learning, pp. 157 163 San Mateo, CA. Morgan Kaufman.
    Google ScholarLocate open access versionFindings
  • Preist, C., Bartolini, C., & Phillips, I. 2001. Algorithm design for agents which participate in multiple simultaneous auctions. In Agent Mediated Electronic Commerce III LNAI, pp. 139 154. Springer-Verlag, Berlin.
    Google ScholarLocate open access versionFindings
  • Rodriguez-Aguilar, J. A., Martin, F. J., Noriega, P., Garcia, P., & Sierra, C. 2001. Towards a test-bed for trading agents in electronic auction markets. AI Communications. In press. Available at http:.sinera.iiia.csic.es ~pablo pncve.html
    Google ScholarFindings
  • Singh, S., Kearns, M., & Mansour, Y. 2000. Nash convergence of gradient dynamics in general sum games. In Proceedings of the Sixteenth Conference on Uncertainty in Arti cial Intelligence UAI, pp. 541 548.
    Google ScholarLocate open access versionFindings
  • Stone, P. 2000. Layered Learning in Multiagent Systems: A Winning Approach to Robotic Soccer. MIT Press.
    Google ScholarFindings
  • Weber, R. J. 1997. Making more from less: Strategic demand reduction in the FCC spectrum auctions. Journal of Economics and Management Strategy, 63, 529 548.
    Google ScholarLocate open access versionFindings
  • Wellman, M. P., Wurman, P. R., O'Malley, K., Bangera, R., Lin, S.-d., Reeves, D., & Walsh, W. E. 2001. A trading agent competition. IEEE Internet Computing, 52, 43 51.
    Google ScholarLocate open access versionFindings
Your rating :
0

 

Tags
Comments
小科