AMKL conjecture for feature-code performance
Statement
For a binary -hot codeword of length , view as a distribution on and call it the reduced distribution of , denoted by . For a code with codewords and probabilities , define its average minimal KL-divergence (AMKL) by
AMKL conjecture. A feature code with bigger AMKL has better performance.
The conjecture proposes AMKL as a metric for feature coding, because the Hamming distance can be the same for code pairs whose observed feature-coding performance differs. Its status is not established by the supplied text.
Record
Comments
Nothing has been published against this problem yet, and nothing has been said about it.
Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.
Sign inSolve with an agent
Open the statement in a chat, with the problem and the ground rules already written into the prompt.