OpenAI research collection · 4 manuscripts · Source checked 2026-10-19
Determines finite-temperature variational free energies for Gaussian Ising perceptrons with bounded Borel log-potentials and Gaussian spherical perceptrons with bounded continuous potentials, at every positive pattern density. A spherical extension treats bi-orthogonally invariant disorder with compact limiting singular-value distributions and no outliers. At margin −1, the quadratic-penalty spherical model has a sharp feasibility threshold and limiting gap and force laws, with system size, zero temperature, and critical density taken in that order.
Short excerpts below come from OpenAI's abstracts. The full abstract and original PDF preserve the paper's precise scope.
We determine the limiting free energy of the Ising perceptron with independent Gaussian patterns for every bounded Borel log-potential, at every fixed positive temperature and pattern density. We give an explicit variational formula for the limit and prove convergence in expectation and probability.
This exact PDF is not listed in the imported formalization catalogue. Family scope notes may contain further information.
We prove a sharp feasibility threshold and limiting gap and force laws for the spherical perceptron with margin −1 and quadratic penalty. The limits are taken successively in system size, inverse temperature, and density approaching the threshold from above. They agree for Gaussian coordinates and for the equal mixture of…
We prove a sharp feasibility threshold and limiting gap and force laws for the spherical perceptron with margin −1 and quadratic penalty. The limits are taken successively in system size, inverse temperature, and density approaching the threshold from above. They agree for Gaussian coordinates and for the equal mixture of centered Gaussian coordinates with variances 1-ε and 1+ε, for every sufficiently small fixed ε. The contact-removed gap cumulative law and the mean-one force cumulative law satisfy \displaystyle G_J(u)=u^{1-\gamma+o(1)}, F_J(s)=s^{1+θ+o(1)}, as u\downarrow0 and s\downarrow0, with \gamma=(2+θ)^{-1}, 0.4126930< \gamma< 0.4126934, and 0.4231063< θ< 0.4231088. A finite numerical certificate for these exponent intervals, together with its mathematical error bounds, is included.
This exact PDF is not listed in the imported formalization catalogue. Family scope notes may contain further information.
We determine the limiting free energy of a spherical perceptron with a bounded continuous activation and a bi-orthogonally invariant disorder matrix. The singular values may have any compact limiting distribution, provided there are no outliers. We give an explicit variational formula for this limit.
This exact PDF is not listed in the imported formalization catalogue. Family scope notes may contain further information.
We prove an exact variational formula for the limiting pressure of the spherical random perceptron with an arbitrary bounded continuous single-pattern potential. The formula holds at every fixed positive density and inverse temperature, with convergence in expectation and in probability.
This exact PDF is not listed in the imported formalization catalogue. Family scope notes may contain further information.
New mathematical reasoning could help AI tackle harder scientific and computational problems. How reliably these skills transfer beyond mathematics will determine how much they contribute to general intelligence.
Explore the AGI Clock ↗