Auction Design using Value Prediction with Hallucinations
This addresses auction design for sellers in settings with potentially flawed AI predictions, though it appears incremental as it builds on Bayesian mechanism design.
The paper tackles the problem of designing revenue-maximizing auctions when machine learning predictions of buyer valuations may be unreliable (hallucinations), characterizing the optimal auction and showing that for one buyer, the seller posts one of three intuitive prices based on the signal.
We investigate a Bayesian mechanism design problem where a seller seeks to maximize revenue by selling an indivisible good to one of n buyers, incorporating potentially unreliable predictions (signals) of buyers' private values derived from a machine learning model. We propose a framework where these signals are sometimes reflective of buyers' true valuations but other times are hallucinations, which are uncorrelated with the buyers' true valuations. Our main contribution is a characterization of the optimal auction under this framework. Our characterization establishes a near-decomposition of how to treat types above and below the signal. For the one buyer case, the seller's optimal strategy is to post one of three fairly intuitive prices depending on the signal, which we call the "ignore", "follow" and "cap" actions.