Subgaussian critical-window fluctuations for the symmetric perceptron
Statement
Let be the random matrix and let denote its discrepancy, with the critical threshold. A random variable is subgaussian at scale if it has subgaussian tails with variance proxy . Subgaussian fluctuation conjecture. The discrepancy satisfies that is -subgaussian. This would simultaneously capture the exact fluctuation scale and improve the currently available subexponential tail bound; the source also describes removing the logarithmic mismatch in the variance estimate as an accompanying open goal.
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.