ProbXiv
Explorersign in
Problem archiveProblem record

Statement

Let N\mathcal{N} be a feedforward neural network with inputs labeled by the coordinate functions x1,…,xnx_1,\dots,x_n, and let F=F(x1,…,xn)F=F(x_1,\dots,x_n) be smooth near a point p∈Rn\mathbf{p}\in\mathbb{R}^n. Write FxαF_{\mathbf{x}^{\mathbf{\alpha}}} for the corresponding partial derivatives. In either setting covered by the paper's two main theorems, Local PDE and PDI sufficiency conjecture. There exist finitely many non-vanishing conditions Ψi((Fxα)∣α∣≤r)≠0\Psi_i\left(\left(F_{\mathbf{x}^{\mathbf{\alpha}}}\right)_{|\alpha|\leq r}\right)\neq 0, finitely many algebraic PDEs Φj((Fxα)∣α∣≤r)=0\Phi_j\left(\left(F_{\mathbf{x}^{\mathbf{\alpha}}}\right)_{|\alpha|\leq r}\right)=0, and finitely many algebraic PDIs Θk((Fxα)∣α∣≤r)>0\Theta_k\left(\left(F_{\mathbf{x}^{\mathbf{\alpha}}}\right)_{|\alpha|\leq r}\right)>0 such that, for every p\mathbf{p}, the non-vacuous space of smooth function germs near p\mathbf{p} computable by N\mathcal{N} and satisfying the non-vanishing conditions at p\mathbf{p} is characterized by those PDEs and PDIs. This is a local sufficiency conjecture for the necessary differential constraints proved earlier; the required conditions and a general proof remain open.

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.