Speaker: Zur Luria, ETH Title: New bounds for the n-queen's problem Abstract: The famous n-queens problem asks: In how many ways can n nonattacking queens be placed on an n by n chessboard? This question also makes sense on the toroidal chessboard, in which opposite sides of the board are identified. In this setting, the n-queens problem counts the number of perfect matchings in a certain regular hypergraph. We give an extremely general bound for such counting problems, which include Sudoku squares and designs. Our lower bound, which confirms a conjecture of Vardi and Rivin, is based on an algebraic construction which is similar to the algebraic absorbers used by Peter Keevash in his construction of designs.
Mon, 07/05/2018 - 11:00 to 12:30
IIAS, Eilat hall, Feldman bldg, Givat Ram