Sinkhorn Based Associative Memory Retrieval Using Spherical Hellinger Kantorovich Dynamics

• URL: http://arxiv.org/abs/2603.20656v1
• Date: Sat, 21 Mar 2026 05:25:15 GMT
• Title: Sinkhorn Based Associative Memory Retrieval Using Spherical Hellinger Kantorovich Dynamics
• Abstract summary: We propose a dense associative memory for empirical measures.<n>Stored patterns and queries are finitely supported probability measures.<n>We derive retrieval dynamics as a spherical Hellinger Kantorovich gradient flow.
• Score: 3.251449696624754
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We propose a dense associative memory for empirical measures (weighted point clouds). Stored patterns and queries are finitely supported probability measures, and retrieval is defined by minimizing a Hopfield-style log-sum-exp energy built from the debiased Sinkhorn divergence. We derive retrieval dynamics as a spherical Hellinger Kantorovich (SHK) gradient flow, which updates both support locations and weights. Discretizing the flow yields a deterministic algorithm that uses Sinkhorn potentials to compute barycentric transport steps and a multiplicative simplex reweighting. Under local separation and PL-type conditions we prove basin invariance, geometric convergence to a local minimizer, and a bound showing the minimizer remains close to the corresponding stored pattern. Under a random pattern model, we further show that these Sinkhorn basins are disjoint with high probability, implying exponential capacity in the ambient dimension. Experiments on synthetic Gaussian point-cloud memories demonstrate robust recovery from perturbed queries versus a Euclidean Hopfield-type baseline.

Related papers

• On the Equivalence of Stochastic Control and Path Space Formulations for Schrödinger Bridges over Compact Connected Lie Groups [0.1259953341639576]
  We establish the equivalence between the numerically optimal control and path space formulations of the Schrdinger bridge problem (SBP)<n>Using the geometric concepts of horizontal lift and anti-development, we derive a Girsanov-type change-of-measure.<n>We illustrate the equivalence to the torus $mathbbT2$.
  arXiv  Detail & Related papers  (2026-09-12T07:08:38Z)
• Lie-Algebraic Subspace Quantization for Zero-Shot Quantum Learning and Barren-Plateau Mitigation [0.9392167468538464]
  We present a framework for zero-shot classical-to-quantum parameter transfer and manifold-based model merging.<n>We prove a Subspace Quantization Theorem that bounds the reconstruction error by geometric truncation and exact non-unitarity.<n>Experiments on IBM ibm_kobe demonstrate Hellinger fidelity of 0.987 at k-8, while subspace gradients remain resolvable up to 128 physical qubits.
  arXiv  Detail & Related papers  (2026-07-13T07:15:12Z)
• On the Stability of Spherical Hellinger-Kantorovich Flows and Their Implications for Differential Privacy [3.251449696624754]
  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.
  arXiv  Detail & Related papers  (2026-05-22T17:38:20Z)
• Fréchet Regression on the Bures-Wasserstein Manifold [2.3204178451683264]
  Fréchet regression, or conditional Barycenters, is a flexible framework for modeling relationships between co- variables on general metric spaces.<n>In this paper, we provide an algorithm for the conditional barycenter problem that characterizes the regression range of the metric space.<n>We also provide an off-the-shelf method for the use of the algorithms for large-scale setups.
  arXiv  Detail & Related papers  (2026-04-04T03:20:19Z)
• Preconditioned One-Step Generative Modeling for Bayesian Inverse Problems in Function Spaces [1.3709465727733763]
  We propose a machine-learning algorithm for inverse problems in the function-space regime based on one-step generative transport.<n>We show that white-noise references may be admissible at fixed discretization, but they become incompatible with the function-space limit.
  arXiv  Detail & Related papers  (2026-03-16T03:52:28Z)
• New Trends in the Stability of Sinkhorn Semigroups [1.0026496861838448]
  Review article presents Sinkhorn/Gibbs-type semigroup analysis based upon contraction coefficients and Lyapunov-type operator-theoretic techniques.<n>New contraction estimates w.r.t. generalized $$-entropies, as well as weighted total variation norms, Kantorovich criteria and Wasserstein distances.
  arXiv  Detail & Related papers  (2026-01-19T00:46:28Z)
• A Mirror-Descent Algorithm for Computing the Petz-Rényi Capacity of Classical-Quantum Channels [13.098901971644656]
  We study the computation of the capacity of a classical-quantum (c-q) channel for $in(0,1)$.<n>We propose an exponentiated-gradient (mirror descent) that generalizes the Blahut-Arimoto algorithm.
  arXiv  Detail & Related papers  (2026-01-15T16:22:52Z)
• Bayesian Circular Regression with von Mises Quasi-Processes [57.88921637944379]
  In this work we explore a family of expressive and interpretable distributions over circle-valued random functions.<n>For posterior inference, we introduce a new Stratonovich-like augmentation that lends itself to fast Gibbs sampling.<n>We present experiments applying this model to the prediction of wind directions and the percentage of the running gait cycle as a function of joint angles.
  arXiv  Detail & Related papers  (2024-06-19T01:57:21Z)
• On Learning Gaussian Multi-index Models with Gradient Flow [57.170617397894404]
  We study gradient flow on the multi-index regression problem for high-dimensional Gaussian data. We consider a two-timescale algorithm, whereby the low-dimensional link function is learnt with a non-parametric model infinitely faster than the subspace parametrizing the low-rank projection.
  arXiv  Detail & Related papers  (2023-10-30T17:55:28Z)
• The Schr\"odinger Bridge between Gaussian Measures has a Closed Form [101.79851806388699]
  We focus on the dynamic formulation of OT, also known as the Schr"odinger bridge (SB) problem. In this paper, we provide closed-form expressions for SBs between Gaussian measures.
  arXiv  Detail & Related papers  (2022-02-11T15:59:01Z)
• Nonconvex Stochastic Scaled-Gradient Descent and Generalized Eigenvector Problems [98.34292831923335]
  Motivated by the problem of online correlation analysis, we propose the emphStochastic Scaled-Gradient Descent (SSD) algorithm. We bring these ideas together in an application to online correlation analysis, deriving for the first time an optimal one-time-scale algorithm with an explicit rate of local convergence to normality.
  arXiv  Detail & Related papers  (2021-12-29T18:46:52Z)
• Inverting brain grey matter models with likelihood-free inference: a tool for trustable cytoarchitecture measurements [62.997667081978825]
  characterisation of the brain grey matter cytoarchitecture with quantitative sensitivity to soma density and volume remains an unsolved challenge in dMRI. We propose a new forward model, specifically a new system of equations, requiring a few relatively sparse b-shells. We then apply modern tools from Bayesian analysis known as likelihood-free inference (LFI) to invert our proposed model.
  arXiv  Detail & Related papers  (2021-11-15T09:08:27Z)
• Learning Optical Flow from a Few Matches [67.83633948984954]
  We show that the dense correlation volume representation is redundant and accurate flow estimation can be achieved with only a fraction of elements in it. Experiments show that our method can reduce computational cost and memory use significantly, while maintaining high accuracy.
  arXiv  Detail & Related papers  (2021-04-05T21:44:00Z)
• Sinkhorn Barycenter via Functional Gradient Descent [125.89871274202439]
  We consider the problem of computing the barycenter of a set of probability distributions under the Sinkhorn divergence. This problem has recently found applications across various domains, including graphics, learning, and vision. We develop a novel functional gradient descent method named Sinkhorn Descent.
  arXiv  Detail & Related papers  (2020-07-20T20:16:47Z)

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.