Uploaded June 2022 | Updated November 2026, 1 day ago
Suppose you form a random word of length n, where the letters are chosen i.i.d. from a probability distribution p = (p_1, ..., p_d) (probabilities in descending order). Let lambda be the Young diagram resulting from passing the word through the RSK process. We expect the normalized row lengths (lambda_1/n, ..., lambda_d/n) to be close to (p_1, ..., p_d). But how close? Does a Central Limit Theorem hold? Is a Central Limit Theorem even good enough?
Opening tune is C-Mone rapping with The Streets.
Thumbnail İznik ceramic from Gallery 460 of The Met.
Suppose you form a random word of length n, where the letters are chosen i.i.d. from a probability distribution p = (p_1, ..., p_d) (probabilities in descending order). Let lambda be the Young diagram resulting from passing the word through the RSK process. We expect the normalized row lengths (lambda_1/n, ..., lambda_d/n) to be close to (p_1, ..., p_d). But how close? Does a Central Limit Theorem hold? Is a Central Limit Theorem even good enough?
Opening tune is C-Mone rapping with The Streets.
Thumbnail İznik ceramic from Gallery 460 of The Met.










