Bipartite Matching Is in NC
4 days ago
- #Parallel Algorithms
- #Bipartite Matching
- #Computational Complexity
- A recent paper claims that the Bipartite Matching problem is in the complexity class NC, solving a long-standing open problem since the 1980s.
- The result derandomizes the Mulmuley-Vazirani-Vazirani algorithm, allowing deterministic polylogarithmic time with parallel processing.
- Bipartite Matching involves pairing n men and n women based on preferences, solvable in polynomial time but now potentially in NC.
- The author mentions a political note about AI regulation, endorsing Alex Bores in a NYC primary due to his stance on AI safety.
- Comments discuss technical aspects, generalizations to matroids, other randomized algorithms needing derandomization, and political reactions.