Answers are grounded in this paper and cite their sources.
We present four new efficient algorithms for random generation of fuzzy measures based on random linear extensions of Boolean lattices. We verify experimentally the uniformity of the resulting distributions and design various informative criteria of uniformity. The algorithms involve Karzanov and Khachiyan random walks but with much shorter mixing times to achieve uniformity. We calculated some important statistical quantities, such as the centroid of the polytope of fuzzy measures, and found patterns for their behaviour in higher dimensions. By matching these patterns we developed random generation algorithms with much higher efficiency. We confirmed empirically the uniformity of the samples.
No takes yet. Share an insight, caveat, or question.
Beliakov et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: