We study the empirical behavior of trading agents participating in the Ad-Auction game of the Trading Agent Competition (TAC-AA). Aiming to understand the applicability of optimal trading strategies in synthesized environments to real-life settings, we investigate the robustness of the agents to deviations from the game's specified environment. Our results indicate that most agents, especially the top-scoring ones, are surprisingly robust. In addition, using the game logs, we derive for each agent a strategic fingerprint and show that it almost uniquely identifies it. Finally, we show that although the Machine Learning modeling in TAC-AA is inherently inaccurate, further improvement in modeling accuracy is likely to have only a limited contribution to the overall performance of TAC-AA agents.
[1]
Michael P. Wellman,et al.
Designing an Ad Auctions Game for the Trading Agent Competition
,
2009,
AMEC/TADA.
[2]
Peter Stone,et al.
A particle filter for bid estimation in ad auctions with periodic ranking observations
,
2011,
AAMAS.
[3]
Victor Naroditskiy,et al.
A Knapsack-Based Approach to Bidding in Ad Auctions
,
2010,
ECAI.
[4]
Michael P. Wellman,et al.
Strategy and mechanism lessons from the first ad auctions trading agent competition
,
2010,
EC '10.
[5]
Peter Stone,et al.
TacTex09: a champion bidding agent for ad auctions
,
2010,
AAMAS.