stat.ML · 2026-10-19 · No. 148

Machine Learning, 2026-10-19.

11 new papers in stat.ML. Titles, authors, abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →

01 — The papers

11 entries
  1. 01

    ENFR: Evidence Networks with Flow-Matched References for Amortized Simulation-Based Model Inference

    Kai Lehman, Jameson Dong, Carolina Cuesta-Lazaro, Niall Jeffrey, Benjamin D. Wandelt, Francisco Villaescusa-Navarro,...

    stat.ML · astro-ph.CO · cs.LG

    The evidence, $\mathcal{Z} = p(x | \mathcal{M})$, of a data-generating model $\mathcal{M}$ is a powerful tool in Bayesian frameworks: it enables model inference, model misspecification tests, and goodness-of-fit assessment as an alternative to standard discrepancy statistics, e.g. $χ^2$. While versatile, the evidence is notoriously difficult to compute, even when the likelihood is available. In both likelihood-based and simulation-based...

    arxiv.org/abs/2610.23825 · PDF

  2. 02

    Learning without relearning physical symmetry: an SO(3)-equivariant convolutional network (CNN) on the sphere

    Joshua Williamson, Niall Jeffrey, Lorne Whiteway

    stat.ML · astro-ph.CO · cs.LG

    Many of the most important datasets in the physical sciences live natively on the sphere, and these data often carry no preferred position or orientation. Using methods that do not reflect this symmetry leads to far higher simulation requirements, which can be catastrophic for scientific analyses (forcing reliance on cheap, imperfect models that bias results). We present a novel spherical implementation of a gauge equivariant mesh...

    arxiv.org/abs/2610.23821 · PDF

  3. 03

    Task-aware Kernel Stein certificates for Rényi differential privacy

    Leoni Carla Wirth, Tara Trautwein, Gesine Reinert

    stat.ML · cs.CR · cs.LG

    Rènyi differential privacy, based on Rènyi divergence, has become a notion of choice for privacy guarantees. However, outside the Gaussian setting, Rènyi divergence can be difficult to compute. In particular, Rènyi differential privacy guarantees that exploit task-specific information, as in Rènyi Pufferfish privacy, can be challenging. Here we use kernel Stein discrepancies based on escort distributions to provide certificates for Rènyi...

    arxiv.org/abs/2610.23798 · PDF

  4. 04

    Parametric Resonance in Primal-Dual Learning for Slowly Non-Stationary Constrained MDPs

    Maxime Baelde

    stat.ML · cs.LG · eess.SY · math.OC

    Primal-dual methods for constrained Markov decision processes are analyzed under a two-timescale assumption, with non-stationarity entering through two channels. In the additive channel, where the constraint level oscillates, the dual oscillation saturates at a floor set by the moving optimum and is amplified linearly by the timescale ratio. In the parametric channel, where the constraint sensitivity oscillates, a coherent modulation carries...

    arxiv.org/abs/2610.23788 · PDF

  5. 05

    Mesoscopic learning dynamics of extensive-width neural networks

    Shu Zhou, Jean Barbier, Sebastian Goldt

    stat.ML · cond-mat.dis-nn · cond-mat.stat-mech · cs.LG

    Understanding how stochastic gradient descent learns internal representations in wide neural networks trained on rich target functions is a central problem in machine learning. We study this question in two-layer neural networks learning target functions represented by networks of the same architecture, with arbitrary nonlinearity, and with network width and input dimension growing proportionally. In this regime, standard approaches for...

    arxiv.org/abs/2610.23779 · PDF

  6. 06

    A general recipe for data-adaptive anytime-valid envelopes for iterative algorithms

    Soham Bonnerjee, Aaditya Ramdas

    stat.ML · cs.LG · math.OC · stat.ME

    Most convergence guarantees for iterative stochastic algorithms are stated at a fixed iteration count and do not justify stopping at a data-dependent time. Recent work has constructed anytime-valid envelopes that do, but deterministic envelopes are typically too conservative for practical stopping, and existing data-adaptive ones are derived case-by-case, each coupling a specific algorithm to a specific noise model and boundary construction....

    arxiv.org/abs/2610.23570 · PDF

  7. 07

    Adaptive conformal prediction on the Grassmannian

    Dimitrios Loukrezis, Zachary J. Grey, Dimitris G. Giovanis

    stat.ML · cs.LG

    We introduce a conformal prediction framework for regression with Grassmann-valued responses. The proposed method constructs prediction regions in the form of metric balls directly on the Grassmann manifold, using basis-invariant geometric distances as non-conformity measures and scaling the corresponding non-conformity scores through a learned prediction difficulty estimator. This yields locally adaptive, input-dependent prediction regions...

    arxiv.org/abs/2610.23508 · PDF

  8. 08

    Towards optimal selection schedules for online active learning

    Hugo Cui, Francesca Mignacco, Francesco Mori

    stat.ML · cond-mat.dis-nn · cs.LG

    We consider the problem of online linear classification, where a stream of unlabeled samples is available, and annotation is costly. We investigate the optimal query policy, that prescribes whether and when a sample should be labeled and used for a training step, or instead discarded, so as to minimize a given objective. We focus on the realizable case, under Gaussian design, in the high-dimensional regime where data dimension and number of...

    arxiv.org/abs/2610.23477 · PDF

  9. 09

    The Loss or the Architecture? What Matters in Survival Analysis with Competing Risks

    Julie Alberge, Gaël Varoquaux, Judith Abécassis

    stat.ML · cs.LG

    Often, \emph{eg} in healthcare and predictive maintenance, what matters is which of several competing causes will strike first, and with what probability: survival analysis with competing risks. Corresponding deep learning methods are developed as bundles: a new architecture together with a new training loss, tested on one to five datasets against a handful of baselines. Thus, a state-of-the-art result cannot be attributed to the...

    arxiv.org/abs/2610.23322 · PDF

  10. 10

    Information-Geometric Blind Spots of Conformal Test Martingales

    Johan Hallberg Szabadváry

    stat.ML · cs.LG · math.ST

    Conformal test martingales (CTMs) are e-processes for sequentially testing exchangeability via p-values produced by a conformity function $A$. However, any distribution shift that preserves the push-forward $P\circ A^{-1}$ leaves the p-value distribution invariant, rendering it invisible to every CTM based on $A$. We formalise this blind spot through the \emph{$A$-equivalence class} $[P]_A$ and show that, for any target distribution $T$,...

    arxiv.org/abs/2610.23131 · PDF

  11. 11

    Distinguishability and Recovery for a Doubly Sparse Spiked Wigner Model

    Ioana Dumitriu, JD Flynn, Zhichao Wang

    stat.ML · cs.LG

    We study distinguishability and recovery of a sparse rank-one signal in a spiked Wigner model with independently sparse noise through a variety of algorithms and parameter regimes. The signal and noise have separate sparsity parameters, while the signal-to-noise ratio (SNR) remains of constant order as the matrix size grows. For sub-Gaussian noise and spikes whose nonzero coordinates are bounded away from zero, a leave-one-out quadratic...

    arxiv.org/abs/2610.23033 · PDF

This edition is part of The Daily Abstract — stat.ML archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.

Colophon Set in Georgia, with system sans for interface chrome and a monospaced stack for code and paper identifiers. Sole accent: amber #D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.