Figure 3
From: Designing all-pay auctions using deep learning and multi-agent simulation
Bidding strategy CDFs. (a) shows the CDFs for each of five different contest designs assuming the bidder plays the Nash distribution. The first prize is listed in the legend; the second prize equals one minus the first; no prize is given to the third bidder. The remaining plots (b–f) compare the Nash CDF with the CDF learned by Fictitious Self-Play for different first prize amounts.
