Learning Revenue-Maximizing Auctions With Differentiable Matching
This addresses auction design challenges for economists and AI practitioners, offering a novel method for specific bottleneck scenarios.
The paper tackles the problem of learning revenue-maximizing auctions with incentive compatibility by proposing a new architecture that uses differentiable bipartite matching via the Sinkhorn algorithm, enabling learning in settings without free disposal where previous methods like RegretNet failed. It successfully recovers known optimal mechanisms and achieves high-revenue, low-regret results in larger, unknown settings.
We propose a new architecture to approximately learn incentive compatible, revenue-maximizing auctions from sampled valuations. Our architecture uses the Sinkhorn algorithm to perform a differentiable bipartite matching which allows the network to learn strategyproof revenue-maximizing mechanisms in settings not learnable by the previous RegretNet architecture. In particular, our architecture is able to learn mechanisms in settings without free disposal where each bidder must be allocated exactly some number of items. In experiments, we show our approach successfully recovers multiple known optimal mechanisms and high-revenue, low-regret mechanisms in larger settings where the optimal mechanism is unknown.