Diffusion posterior samplers produce biased outputs that can be expressed as an Ornstein-Uhlenbeck path expectation via a surrogate Gaussian path and Feynman-Kac representation, with STSL flattening the spatially varying bias term.
Title resolution pending
14 Pith papers cite this work. Polarity classification is still indexing.
14
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 14roles
background 1polarities
background 1representative citing papers
The second term in the spectral expansion of the expected LQG heat trace as t to 0 is governed by the KPZ exponent.
The value function is the unique minimizer of a functional whose Euler-Lagrange equation matches the HJB equation, with a verification theorem giving optimal policies via change of measure and BSDE uniqueness.
Martingale Neural Operator uses Doob-Meyer factorization to output mean and low-rank covariance for stochastic PDE terminal laws, achieving large Wasserstein reductions versus diffusion baselines on tested SPDEs.
The Sinkhorn treatment effect is a new entropic optimal transport measure of divergence between counterfactual distributions that admits first- and second-order pathwise differentiability, debiased estimators, and asymptotically valid tests for distributional treatment effects.
Generalized bridges with constraints solve Schrödinger problems, enabling broader financial equilibrium models with frictions and proving convergence of trading-cost equilibria to the classical Kyle model.
Derives Õ(d β² A² / ε⁴) oracle complexity for AIS estimating normalizing constant Z to relative error ε and introduces reverse diffusion sampler for geometric paths with large action.
Existence is proved of a slowed-down sticky Brownian motion that induces a MAXCUT rounding attaining the Goemans-Williamson approximation ratio.
A stochastic-geometric model of solution-space topology under Adam derives explicit scaling laws for grokking transition time as a function of learning rate, batch size, and L2 coefficient.
Defines resilience evaluation D^ρ π as the L1-limit of scaled dynamic risk measure applied to process increments, and derives its dual representation as worst-case conditional expectation of an effective drift when ρ arises from BSDEs with Lipschitz or quadratic drivers.
Characterizes duals of white-noise-driven continuous stochastic flows by explicit SDEs and introduces a self-dual polynomially self-repelling flow model.
Global existence of H¹ martingale solutions to the stochastic Camassa-Holm equation is shown via viscous Galerkin approximations, tightness, and Skorokhod-Jakubowski representations.
Proves almost sure continuous dependence of the solution map on initial data in H^s (s>3/2) and existence of non-unique invariant measures for the Camassa-Holm equation with linear multiplicative noise.
Adapts hybrid LSMC-PDE framework to the GDMR model for Bermudan option pricing and reports lower errors than plain LSMC in numerical experiments with low to moderate simulation paths.
citing papers explorer
-
Sinkhorn Treatment Effects: A Causal Optimal Transport Measure
The Sinkhorn treatment effect is a new entropic optimal transport measure of divergence between counterfactual distributions that admits first- and second-order pathwise differentiability, debiased estimators, and asymptotically valid tests for distributional treatment effects.
-
Complexity Analysis of Normalizing Constant Estimation: from Jarzynski Equality to Annealed Importance Sampling and beyond
Derives Õ(d β² A² / ε⁴) oracle complexity for AIS estimating normalizing constant Z to relative error ε and introduces reverse diffusion sampler for geometric paths with large action.
-
A Stochastic--Geometric Theory of Scaling Laws in Grokking
A stochastic-geometric model of solution-space topology under Adam derives explicit scaling laws for grokking transition time as a function of learning rate, batch size, and L2 coefficient.