Multi-Unit Auctions for Allocating Chance-Constrained Resources

Sharing scarce resources is a key challenge in multi-agent interaction, especially when individual agents are uncertain about their future consumption. We present new auction mechanism for preallocating multi-unit among agents, while limiting the chance of resource violations. By planning constraint, we strike balance between worst-case approaches, which under-utilise resources, and expected-case lack formal guarantees. also an algorithm that allows to generate bids via multi-objective reasoning, then submitted auction. discuss how can be extended non-cooperative scenarios. Finally, demonstrate empirically our outperforms state-of-the-art techniques chance-constrained allocation complex settings with up hundreds agents.