ProbXiv
Explorersign in
Problem archiveProblem record

Statement

Let p∈(2,∞)p\in(2,\infty). For w∈(−1,1)∖{0}w\in(-1,1)\setminus\{0\}, consider the variational problem

inf⁡ν∈P(R){I∞,μpan(w)+H(ν∣μ2)+12(1−m2(ν))},\inf_{\nu\in\mathcal{P}(\mathbb{R})}\left\{\mathbb{I}^{\mathsf{an}}_{\infty,\mu_p}(w)+H(\nu\mid\mu_2)+\frac{1}{2}(1-m_2(\nu))\right\},

with the minimizer understood as in. For γ∈Tp\gamma\in\mathcal{T}_p and w∈DIγan∖{0}w\in D_{\mathbb{I}^{\mathsf{an}}_{\gamma}}\setminus\{0\}, consider analogously the variational problem in.

Non-Gaussian minimizer conjecture. For every p∈(2,∞)p\in(2,\infty) and w∈(−1,1)∖{0}w\in(-1,1)\setminus\{0\}, the minimizer in is not μ2\mu_2. Similarly, for every γ∈Tp\gamma\in\mathcal{T}_p and w∈DIγan∖{0}w\in D_{\mathbb{I}^{\mathsf{an}}_{\gamma}}\setminus\{0\}, the minimizer in is not μ2\mu_2. Consequently, except at w=0w=0, the annealed rate function lies strictly below the quenched rate function.

This conjecture would extend the preceding proposition from the Gaussian setting to all p∈(2,∞)p\in(2,\infty) and to products of measures other than γ=μ∞\gamma=\mu_\infty. Its resolution would clarify when the annealed and quenched rate functions differ strictly away from the origin.

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 in

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.