On the Stability of Spherical Hellinger-Kantorovich Flows and Their Implications for Differential Privacy

• URL: http://arxiv.org/abs/2605.23879v1
• Date: Fri, 22 May 2026 17:38:20 GMT
• Title: On the Stability of Spherical Hellinger-Kantorovich Flows and Their Implications for Differential Privacy
• Abstract summary: We develop a perturbation theory for SHK gradient flows and quantify how potential discrepancies propagate over time.<n>A uniform bound yields dimension-free, pointwise control of the log-likelihood ratio and Rényi divergence.<n>We apply these results to approximate sampling for the exponential mechanism in differential privacy.
• Score: 3.251449696624754
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Gradient-flow sampling interprets a Gibbs distribution as the minimizer of an energy functional over probability measures and generates dynamics converging to this target. Under spherical Hellinger-Kantorovich (SHK) geometry, the flow couples transport and reaction and coincides with birth-death Langevin dynamics. In this work, we develop a perturbation theory for SHK gradient flows. For two potentials $V$ and $V^{\prime}$, we compare the associated flows from a common initialization and quantify how potential discrepancies propagate over time. A uniform perturbation bound yields dimension-free, pointwise control of the log-likelihood ratio and Rényi divergence, while additional structure allows us to derive bounds for the KL divergence as well. We apply these results to approximate sampling for the exponential mechanism in differential privacy. The likelihood-ratio control provides explicit time-dependent Pure-DP guarantees for SHK-based samplers, while the KL bound yields Approximate-DP certificates via hockey-stick divergence. We also derive a utility bound separating intrinsic exponential-mechanism suboptimality from finite-time sampling error.

Related papers

• LC-GRPO: Bridging Train-Inference Gap for Flow-Based GRPO with Langevin Correction [59.32834240526981]
  Flow-based generative models are typically sampled by solving a deterministic ordinary differential equation (ODE)<n>Existing GRPO methods for flow models replace the Euler-time ODE with aotropic differential equation (SDE) during training.<n>We introduce LC-GRPO, a flow-based GRPO framework with Langevin correction targeting the marginal distribution at the resulting timestep.
  arXiv  Detail & Related papers  (2026-08-06T04:47:57Z)
• Adaptive MPPI with Online Disturbance Covariance Estimation: Provable Stability Tightening via Spatial Smoothing [0.0]
  We study Model Predictive Path Integral (MPPI) control for nonlinear systems with additive process disturbances.<n>A mismatched disturbance covariance produces a persistent penalty in closed-loop stability certificates.
  arXiv  Detail & Related papers  (2026-07-09T21:12:40Z)
• Disentangling Continuous-Time Latent Dynamics: Identifiability of Latent SDEs via Diffusion Shifts [60.95873409673775]
  identifiability in continuous-time latent differential equation models remains largely open.<n>We study additive-noise latent SDEs observed through an unknown nonlinear diffeomorphism.<n>We propose a two-stage estimator for latent disentanglement and optional graph recovery.
  arXiv  Detail & Related papers  (2026-06-26T16:18:28Z)
• A Quantitative Approximation Framework for Flow Distillation in Diffusion Models [9.704954388992892]
  Local approximation errors can be strongly amplified in low-noise multimodal regimes.<n>We show that deep residual compositions efficiently approximate the long-horizon transport.<n>Experiments support the prediction and reduce end-to-end relative MSE by up to 51.9% with 8 segments compared with uniform grids.
  arXiv  Detail & Related papers  (2026-06-02T16:00:43Z)
• Why SGD is not Brownian Motion: A New Perspective on Stochastic Dynamics [92.39053980710702]
  Gradient Descent (SGD) is commonly modeled as a Langevin process, assuming that minibatch noise acts as Brownian motion.<n>This approximation relies on a continuous-time limit and a sqrt(eta) noise scaling that does not match the discrete SGD update at finite learning rate.<n>We propose an alternative formulation of SGD as deterministic dynamics in a fluctuating loss landscape induced by minibatch sampling.
  arXiv  Detail & Related papers  (2026-05-21T15:50:40Z)
• Surprisal-Rényi Free Energy [3.0518869830825293]
  We introduce the Surprisal-Rényi Free Energy (SRFE), a log-moment-based functional of the likelihood ratio that lies outside the class of $f$-divergences.<n>We show that SRFE recovers forward and reverse Kullback-Leibler divergences as singular endpoint limits and derive local expansions around both limits in which the variance of the log-likelihood ratio appears as a first-order correction.
  arXiv  Detail & Related papers  (2026-03-03T16:42:08Z)
• Parallel Complex Diffusion for Scalable Time Series Generation [50.01609741902786]
  PaCoDi is a spectral-native architecture that decouples generative modeling in the frequency domain.<n>We show that PaCoDi outperforms existing baselines in both generation quality and inference speed.
  arXiv  Detail & Related papers  (2026-02-10T14:31:53Z)
• Particle Dynamics for Latent-Variable Energy-Based Models [12.84928511163926]
  Latent-variable energy-based models (LVEBMs) assign a single normalized energy to joint pairs of observed data and latent variables.<n>We recast maximum-likelihood training as a saddle problem over distributions on the latent and joint gradients.<n>We prove existence and convergence under standard smoothness and dissipativity assumptions, with decay rates in KL divergence and Wasserstein-2 distance.
  arXiv  Detail & Related papers  (2025-10-17T09:04:49Z)
• The Effect of Stochasticity in Score-Based Diffusion Sampling: a KL Divergence Analysis [0.0]
  We study the effect of divergenceity on the generation process through bounds on the Kullback-Leibler (KL)<n>Our main results apply to linear forward SDEs with additive noise and Lipschitz-continuous score functions.
  arXiv  Detail & Related papers  (2025-06-13T01:01:07Z)
• Sampling with Adaptive Variance for Multimodal Distributions [14.121491356732188]
  We propose and analyze a class of distributions sampling algorithms for a bounded domain. We show that a derivative-free version can be used for sampling without information on the Gibbs potential.
  arXiv  Detail & Related papers  (2024-11-20T22:05:47Z)
• Rectified Diffusion Guidance for Conditional Generation [94.83538269086613]
  We revisit the theory behind CFG and rigorously confirm that the improper combination coefficients (textiti.e.) brings about expectation shift the generative distribution.<n>We show that our approach enjoys a textbftextitform solution given the strength.<n> Empirical evidence on real-world data demonstrate the compatibility of our design with existing state-of-the-art diffusion models.
  arXiv  Detail & Related papers  (2024-10-24T13:41:32Z)
• Your Absorbing Discrete Diffusion Secretly Models the Conditional Distributions of Clean Data [55.54827581105283]
  We show that the concrete score in absorbing diffusion can be expressed as conditional probabilities of clean data.<n>We propose a dedicated diffusion model without time-condition that characterizes the time-independent conditional probabilities.<n>Our models achieve SOTA performance among diffusion models on 5 zero-shot language modeling benchmarks.
  arXiv  Detail & Related papers  (2024-06-06T04:22:11Z)
• Adaptive Annealed Importance Sampling with Constant Rate Progress [68.8204255655161]
  Annealed Importance Sampling (AIS) synthesizes weighted samples from an intractable distribution. We propose the Constant Rate AIS algorithm and its efficient implementation for $alpha$-divergences.
  arXiv  Detail & Related papers  (2023-06-27T08:15:28Z)
• On the Practicality of Differential Privacy in Federated Learning by Tuning Iteration Times [51.61278695776151]
  Federated Learning (FL) is well known for its privacy protection when training machine learning models among distributed clients collaboratively. Recent studies have pointed out that the naive FL is susceptible to gradient leakage attacks. Differential Privacy (DP) emerges as a promising countermeasure to defend against gradient leakage attacks.
  arXiv  Detail & Related papers  (2021-01-11T19:43:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.