Perhaps look into Shapley values as well?

Interesting, yes. My understanding is Shapley is more about allocating a fixed surplus based on marginal contributions, whereas I’m trying to find the agreement itself given inferred preferences. But definitely related territory.