arXiv Machine Learning

Variational Mixtures and Multi-Marginal Flow Matching: Advancing Statistical Inference with Biological Applications

arXiv Machine Learning
Jun 19

Variational Consensus Monte Carlo for Bayesian Mixture

arXiv:2606. 19643v1 Announce Type: cross Abstract: Motivated by the privacy, sensitivity and sharing limitations of health data, we present a comprehensive pipeline for inference of Bayesian mixture models within a federated learning setting, i.

By Julie Fendler, Francesca L. Crowe, Tom Marshall, Sylvia Richardson, Paul D. W. Kirk
arXiv Machine Learning
Oct 28

Cumulative Marginal Matching for One-Step Diffusion

The paper introduces Cumulative Marginal Matching (CMM), a new objective for training few‑step diffusion and flow‑matching models. CMM treats flow‑matching as a sequence of intermediate marginal distributions and directly optimizes the cumulative Wasserstein‑2 discrepancy between the base and few‑step models across the entire trajectory. It derives a tractable surrogate for this objective, shows that Shortcut‑CSL is a first‑order approximation of CMM, and reports improved generation quality on standard benchmarks under the same training budget.

By Paribesh Regmi, Rui Li
arXiv Machine Learning
Jun 9

Dendrograms of Mixing Measures for Softmax-Gated Gaussian Mixture of Experts: Consistency Without Model Sweeps

arXiv:2510. 12744v2 Announce Type: replace-cross Abstract: We develop a unified statistical framework for softmax-gated Gaussian mixture of experts (SGMoE) that addresses three long-standing obstacles in parameter estimation and model selection: (i) non-identifiability of gating parameters up to common translations, (ii) intrinsic gate-expert interactions that induce coupled differential relations in the likelihood, and (iii) the tight numerator-denominator coupling in the softmax-induced conditional density.

By Do Tien Hai, Trung Nguyen Mai, TrungTin Nguyen, Nhat Ho, Binh T. Nguyen, Christopher Drovandi
arXiv Machine Learning
Jun 16

Amortized mean-shift interacting particles

arXiv:2606. 15871v1 Announce Type: cross Abstract: Bayesian inference for inverse problems is run to evaluate integrals -- posterior expectations, tail probabilities, and risks -- across a stream of observations.

By Ali Siahkoohi
arXiv Statistics ML
Nov 9

ELPAC: Endpoint-Anchored Latent Progression with Stage-Varying Multimodal Coordination

ELPAC is a Bayesian framework that jointly infers subject staging, progression subtypes, and stage‑varying multimodal coordination from partially observed aging cohort data. It integrates subtype trajectories backward from clinically reliable endpoint groups using biologically structured velocity components, and then factorizes residual multimodal variation into sparse coordination axes that reveal which multimodal patterns are active in each subtype and when they emerge. The method employs scalable amortized variational inference, enabling accurate latent‑structure recovery and well‑calibrated predictions on both synthetic data and a harmonized Alzheimer’s cohort.

By Shengxian Ding, Yifei Zhang, Xinyuan Tian, Yuanshuang Guo, Yize Zhao
arXiv Machine Learning
Aug 5

Information-Geometric Forward Policy Training in GFlowNets

arXiv:2608. 03967v1 Announce Type: cross Abstract: Generative Flow Networks (GFlowNets) have emerged as a flexible framework for amortised inference over discrete and mixed discrete-continuous objects, requiring only an unnormalised target density specified through a reward.

By Yordan Raykov, Rodrigo Veiga