New Machine Learning (Statistics) (stat.ML) papers, Wednesday 21 October 2026
40 papers announced on arXiv. Get them ranked for you, free →
Quadratic Sample Complexity for Strictly Replicable Realizable Learning with Unknown Marginals
AI summaryThis paper establishes quadratic sample complexity bounds for strictly replicable learning under unknown marginal distributions. The main result proves a tight $\Theta(d^2/(\varepsilon\rho^2))$ sample complexity for learning the Boolean class $d$ under strict replicability requirements.
Dynamic Dual-Feedback Conformal Inference for Time Series Forecasting
AI summaryThis paper introduces Dynamic Dual-Feedback Conformal Inference, a method for time series uncertainty quantification that combines actual miscoverage feedback with estimated conformal thresholds. The approach achieves more stable and narrower prediction intervals while maintaining target coverage.
Robust Latent Variable Graphical Models: Exact Recovery and Fundamental Limits
AI summaryConvex estimator recovers a Gaussian precision matrix decomposed into a sparse minus low‑rank part using measure‑transformed covariance and a nuclear‑norm penalty; it achieves exact edge, sign, rank, and inertia recovery with optimal sample complexity and robustness to Huber outliers.
Reliable Estimation of Differential Privacy under Selection Bias
AI summaryThis paper identifies and corrects a selection bias in membership inference attacks used for differential privacy audits. The proposed statistical corrections yield valid privacy lower bounds that respect theoretical upper bounds at the expected confidence levels.
Gaussian Approximation of Besov Functions for Logistic Classification on Manifolds
AI summaryGaussian kernel logistic classifiers on a compact manifold achieve near‑optimal excess risk without assuming a bounded Bayes function. The authors construct bias‑canceling Gaussian approximants that respect intrinsic dimension, yielding logistic excess‑risk rates matching minimax bounds for bounded densities.
Hilbert-Scale Stability of Posterior Diffusion Priors for Ill-Posed Signal Recovery
AI summaryThis paper establishes a Hilbert-scale stability theory for posterior diffusion priors in ill-posed inverse problems. The main result provides non-asymptotic recovery bounds that define an explicit regularity threshold where score errors remain controlled under smoothing forward operators.
Modular Bayesian Inference for Partially Misspecified Mixtures with Uncertain Class Labels
AI summaryModular Bayesian inference decouples uncertain class labels from downstream parameter estimation by fixing class probabilities from a classifier, then fitting protected models per label allocation. When class probabilities are calibrated, the posterior of the target parameters remains unbiased, as shown in simulations and supernova cosmology.
Learned Control Variates for Rare Grid-Outage Estimation at Electric Vehicle Charging Stations
AI summaryLearned control variates using a classifier on cheap features cut Monte Carlo variance by up to 60% for rare EV‑charging feeder overloads, reducing estimated blackout days by a third in Surabaya’s network compared to plain simulation.
Diffusion Score Estimation with Shallow Neural Networks: Regularity, Approximation, and Finite-Sample Rates
AI summaryThe paper shows that shallow ReLU networks trained by denoising score matching can estimate Ornstein–Uhlenbeck diffusion scores with high-probability risk $\widetilde O(n^{-(d+3)/(2d+3)})$, yielding a total-variation error $\widetilde O(n^{-(d+3)/(4d+6)})$ for reverse-process sampling.
Solving the Ho--Nguyen Conjecture for Gaussian Mixtures: Applications to Mixtures of Experts
AI summaryShows that for any over‑fitted Gaussian mixture of size ≥ 2 the cancellation polynomial has exact order 2M, proving Ho–Nguyen’s conjecture. This yields the sharp learning exponent and local minimax rates, and clarifies symmetry‑based rate limits in mixtures of experts.
VampPrior Unleashed
Deep latent variable models (DLVMs) underpin many modern analysis pipelines, for example in single-cell genomics, yet their prior distributions have received markedly little attention. The VampPrior approximates the optimal prior (the aggregate posterior) and often attains high…
A plug-in estimator of the non-iterative conditional expectation g-formula using long short-term memory models
AI summaryThis paper introduces an LSTM-based plug-in estimator for the NICE g-formula to estimate causal effects in longitudinal data with time-varying confounding. Experiments demonstrate that the LSTM approach reduces bias compared to standard parametric estimators while maintaining nominal confidence interval coverage.
Communication Complexity of Universal Features
AI summaryThis paper studies the communication complexity of universal data representations for arbitrary function classes. The authors prove that an encoder can compress data into $O(d^\log(1/\varepsilon))$ bits—where $d^$ is the dual VC dimension—achieving worst-case optimality across all representation tasks.
A Bayesian machine learning approach to direct and spillover effects in cluster-randomized trials with noncompliance
AI summaryThis paper introduces a Bayesian machine learning method combining additive regression trees with random effects to estimate direct and spillover effects in cluster-randomized trials with noncompliance. Simulations and an empirical application demonstrate that the approach reduces bias under model misspecification and improves effect decomposition.
Beyond Scalar Leaderboards: Adaptive Sampling for Pareto Frontier Identification in Multi-Objective LLM Arenas
The evaluation of large language models (LLMs) is inherently multi-objective, as different models may offer different tradeoffs across performance dimensions such as math, code, and writing. Because such evaluations are often based on costly and time-consuming human feedback, it…
MiCLA: Mixture-to-Constituent Latent Alignment for Polyphonic Audio Representation Learning
AI summaryMiCLA is a self-supervised pre-training framework for polyphonic audio that aligns mixture representations with their constituent clips using bidirectional latent alignment. It achieves state-of-the-art performance on AudioSet fine-tuning and probing evaluations, effectively handling concurrent sound events.
On the Accuracy of Approximate Leave-One-Out with Blockwise Nonsmooth Regularization in High-dimensions
AI summaryThe paper proves that a heuristic ALO method for block‑sparse, non‑smooth regularizers in high‑dimensional settings has an average error of \(O(\log^{3/2}(en)/\sqrt{n})\), showing strong theoretical guarantees and insights into estimator stability.
What Moved the Alarm Score? Causal Attribution with Exact Computation and Scarce-Incident Estimation
Alarms in industrial systems and scientific instruments often demand immediate diagnosis despite scarce incident data. To guide intervention, an explanation must trace how changes in the data-generating process propagate through the causal structure to alter the alarm score,…
Active Pilot Sampling for Multi-fidelity Estimation in Stochastic Optimization
AI summaryThis paper introduces an active pilot-sampling framework for multi-fidelity stochastic optimization that emulates model-output covariances across the design space. The method improves optimization efficiency and performance compared to standard Monte Carlo and baseline pilot-sampling strategies.
Bayesian Neural Networks via Dirichlet Process Priors on the Data-Generating Distribution for Randomized Value Functions in Deep Reinforcement Learning
Bayesian neural networks (BNNs) offer a principled framework for quantifying epistemic uncertainty. However, standard weight-space priors yield complex, high-dimensional neural network posteriors that are computationally prohibitive to sample from, and popular approximate…
From POMDP Theory to Deep RL with Particle Filters
Model-based reinforcement learning in partially observable environments requires inferring hidden states, learning latent dynamics, and planning under the resulting belief. Practical algorithms often solve one aspect of the above pipeline while relying on approximations for the…
Ask to Learn: Query-Based Collaborative Reinforcement Learning
Offline reinforcement learning is fundamentally limited by the coverage of the available data, while collecting additional trajectories through online exploration may be costly, unsafe, or infeasible. We study whether this limitation can be overcome without directly interacting…
Divide-and-Conquer Neural Bayes Estimation for Spatial and Spatiotemporal Point Process Models
AI summaryThe paper introduces a divide‑and‑conquer neural Bayes estimator using transformer models on spatial and spatiotemporal point processes, combining local summaries with attention. It achieves scalable, accurate inference with credible intervals, outperforming existing neural and classical likelihood methods on simulations and whale‑call data.
BARRR: Bidimensional Augmented Reduced Rank Regression for Multi-View Multi-Cohort Data
AI summaryBidimensional Augmented Reduced Rank Regression (baRRR) jointly models covariate-related/unrelated effects and decomposes global, partially shared, and individual components across cohorts and data views. Simulations show improved signal recovery, prediction, and missing data imputation, and TCGA application uncovers mutation-informed multi‑omics patterns across cancers.
Pattern Intelligence: Unsupervised, LLM-Interpreted Discovery of Systematic Underpayment and Denial Patterns in Healthcare Revenue Cycle Management
AI summaryPattern Intelligence clusters claims by adjudication behavior, not payer, to discover systematic underpayment and denial patterns. Using HDBSCAN and LLM-annotated centroids, it identifies actionable clusters, revealing a long tail of high-dollar denials that remain uncaptured by traditional methods.
Privacy Accounting for Empirical Denoisers
AI summaryThe paper proposes a clipped, privacy‑accountable empirical denoiser that allows exact sensitivity calculation for diffusion sampling, enabling DP guarantees for each denoising step. It shows privacy loss peaks during semantic formation, motivating hybrid samplers that switch to a public denoiser, and reports concrete (ε, δ) bounds on CIFAR‑10 and CelebA‑64.
Practical Advantages of Online Conformal Prediction for Data Streams
AI summaryOnline conformal regression methods are compared to a weighted full‑conformal baseline on IID, shifting, and real streams; adapting both the model and the quantile threshold yields better coverage‑width tradeoffs, with the quantile‑tracking approach costing \(O(p^2)\) per update.
Annealed correlation and quenched noise sensitivity in biased deep neural networks
AI summaryAdding independent Gaussian biases of natural scale, the paper proves quenched noise sensitivity still holds for deep fully connected nets with sign activations whenever the perturbed input is asymptotically unlikely to equal the original, while the annealed covariance stays bounded away from zero, driven entirely by output‑mean fluctuations.
Optimal aggregation in multiclass classification
AI summaryThe paper extends aggregation theory from binary to multiclass classification, proving that optimal rates for 0/1 and logistic losses are independent of the number of classes. It shows proper aggregation can degrade with K, while improper probability‑mixing aggregates recover optimal, class‑size‑free rates.
When Does Training on Synthetic Data Provably Work?
AI summaryThis paper analyzes when classifiers trained on synthetic data successfully transfer to real distributions. The authors prove that synthetic training can fail catastrophically when real data falls outside the synthetic support, unless constrained by Lipschitz conditions.
Quantifying the Threat: A Bayesian Probabilistic Framework for Heavy Metal Health Risks among Children in the Densu Basin, Ghana
AI summaryThe authors develop a Bayesian risk‐assessment framework for groundwater in Ghana’s Densu Basin, incorporating parameter uncertainty, co‑exposure, and spatial variability. They find Pb, As, and Cd drive cumulative risk, with ~31 % of sites exceeding a hazard index of 1, revealing highly heterogeneous, uncertainty‑aware risk surfaces.
Near-Optimal Regret Bounds for UCB-Style Gaussian Process Bandits via Data Thinning
AI summaryThis paper introduces a data thinning technique to Gaussian process upper confidence bound algorithms. It proves that this approach yields tighter confidence intervals and achieves near-optimal simple and cumulative regret bounds for GP bandits.
Predicting silent failures of learned inference under restricted access: what measurement physics and prior information explain
AI summaryThis paper investigates predicting silent failures in learned image reconstruction under restricted measurement access. The main result is that while a new score can anticipate these failures, their occurrence is driven primarily by measurement geometry and physics rather than the learned prior.
Where Has the Foreign Signal Gone?
We compare the relation between sentiment noise in the Twitter feed and losses for stocks of domestic and foreign firms. The mean quarterly domestic-foreign difference in the slope of next-quarter losses on measured noise is -1.40 (Newey-West t = -2.99). This slope is negative…
Geometry-Aware Block Selection for Sparse Variational Gaussian Processes
AI summaryThis paper introduces a geometry-aware algorithm to select block partitions for structured scaling matrices in sparse variational Gaussian processes. The method improves predictive performance and tightens the variational lower bound with minimal computational overhead.
Debiased Prior-Data Fitted Networks for Amortized Bayesian Semi-Parametric Inference
AI summaryThis paper analyzes the statistical behavior of prior-data fitted networks (PFNs) for semi-parametric Bayesian inference and introduces efficient influence function-based debiasing strategies. The authors prove that the proposed debiased PFNs achieve optimal convergence rates that match traditional frequentist estimators, establishing a bridge between amortized Bayesian inference and semi-parametric efficiency theory.
Stabilizing Filters for Graph Signal Generation
Generating graph signals is an important task in machine learning, and incorporating the underlying graph into a generative model enables structurally-informed distribution modeling. However, said graph is often noisy or imperfectly estimated, and structural errors can affect…
Confessions of a Masked MDP
AI summaryThis paper studies reinforcement learning in Markov Decision Processes where the set of valid actions is initially unknown and must be learned alongside the optimal policy. The authors propose computationally efficient, minimax-optimal algorithms that actively discover action masks and apply them to transfer learning settings.
Do Neural Models Prefer Their Submodels?
AI summaryThe paper analyzes whether loss minima of neural networks lie within their natural submodels. It shows that shallow MLPs never favor subnetworks with inactive neurons, deep MLPs rarely do, and most self‑attention submodels are not preferred, suggesting geometry alone doesn’t explain submodel preference.
Binary Discrete Diffusion: Candidate-Set Diffusion for Logic and Planning Tasks
AI summaryBinary Discrete Diffusion lets a diffusion model maintain candidate token sets as binary vectors, gradually adding/removing options to represent partial uncertainty. It outperforms masked and uniform baselines on hard planning tasks (Sudoku-Extreme, Boxoban, graph coloring) and remains competitive on language modeling.