Local privacy mechanisms preserve rate-double-robustness, enabling unbiased and semiparametrically efficient inference on target parameters indexed linearly by infinite-dimensional and nonlinearly by low-dimensional components from noisy private data.
Canonical reference
Title resolution pending
Canonical reference. 83% of citing Pith papers cite this work as background.
67
Pith papers citing it
Background
83% of classified citations
citation-role summary
background 5
method 1
citation-polarity summary
representative citing papers
The authors give an efficient non-interactive L-LDP algorithm for SCO achieving excess risk O(sqrt(K/(ε n))) in high privacy and O(sqrt(K/(e^ε n))) in medium privacy, with matching information-theoretic lower bounds for large n.
SCOREBED isolates EIG double intractability in a policy-independent score-matching stage, then trains design policies with a singly intractable gradient estimator, enabling cheap multi-policy selection.
A practical input-to-state stability certificate for Koopman learning control is derived, separating prediction residuals from selected-channel margins and projection residuals.
A multi-scale spectral pipeline using deep learning filament detection automatically identifies 91 oscillatory events in two weeks of 2014 GONG data, recovering known events and finding new ones with periods 20-126 min.
Develops stochastic policies and single-basis-function modification for causal inference on functional treatments, proves asymptotic normality and rate double robustness, and applies to NHANES physical activity and mortality data.
Signed pairwise interaction scores conflate U/R/S; Stochastic Hi-Fi uses interventional masked inference to recover per-feature uniqueness, redundancy, and synergy profiles.
In small-budget RCTs where significance tests decide scale-up, optimal pilot sampling shifts from representative to single homogeneous subpopulation as budget shrinks.
EFE-based planning is formulated as variational free energy minimization with epistemic priors, decomposing into expected plan costs plus a complexity term.
A Bayesian global Fréchet regression method is introduced via a Fréchet Bayes rule that reduces the problem to scalar tasks, allows prior-data interpolation, and remains valid under moment conditions using weak conditional expectations.
Establishes statistical and computational optimality thresholds for common subspace estimation and inference under varying SNR regimes, including an impossibility result for adaptive confidence intervals below strong inference SNR.
Proper EFE-based planning is VFE plus planning and epistemic entropy corrections, realized by channel-reparameterized message passing that captures novelty.
Under clientwise sample-level zCDP, the Fisher information of any fully interactive public federated transcript contracts to a sum of per-client privacy-vs-sample terms, yielding matching minimax rates for mean, linear, and nonparametric regression.
Federated martingale posterior sampling lets clients share data embeddings for central predictive Bayesian sampling, matching centralized performance and improving calibration on MNIST, CIFAR-10, and CIFAR-100.
DCO decouples tuning for efficiency from calibration for coverage in conformal prediction, maintaining marginal guarantees and reducing average set sizes on benchmarks like ImageNet-A and Diabetes.
The PMNLV model extends single-neuron overdispersion to populations via matrix-normal gain priors, showing shared co-variability highest in V1 and declining along the mouse visual hierarchy.
The authors propose an S-MILP framework that optimizes group sequential testing boundaries to achieve faster rejection of the null hypothesis compared to traditional methods while controlling type I and type II errors.
PLACE delivers a closed-form certified classification method for point clouds and graphs based on persistent homology with explicit excess-risk bounds, selection rules, and training-time certificates.
DVISR performs variational inference over symbolic expression trees and constants by training a neural network with the ELBO as reward, recovering true posteriors in simple test cases.
Conditional expectations of latent variables under additive noise admit unique continuous Tweedie representations via a linear functional constructed from the inverse Fourier transform of an explicit tempered distribution.
New theoretical results on estimators and intervals for predicting unseen outcomes in additional samples from discrete distributions, with extensions to grouped incidence data.
Gaussian particles in a linearized Bures-Wasserstein space perform consensus optimization for variational inference and outperform deterministic gradient methods on low-dimensional non-log-concave targets.
Heat-kernel smoothing over weighted points on a compact manifold yields a scale-dependent geometric effective sample size that discounts nearby and duplicate particles.
Introduces autorelevance functions as lag importance measures for time series forecasting using Shapley values and a novel one-step forecast replacement for absent features.
citing papers explorer
-
Private Rate-Double-Robust Inference
Local privacy mechanisms preserve rate-double-robustness, enabling unbiased and semiparametrically efficient inference on target parameters indexed linearly by infinite-dimensional and nonlinearly by low-dimensional components from noisy private data.
-
Convex Optimization with Local Label Differential Privacy: Tight Bounds in All Privacy Regimes
The authors give an efficient non-interactive L-LDP algorithm for SCO achieving excess risk O(sqrt(K/(ε n))) in high privacy and O(sqrt(K/(e^ε n))) in medium privacy, with matching information-theoretic lower bounds for large n.
-
Bayesian Experimental Design via Score Matching
SCOREBED isolates EIG double intractability in a policy-independent score-matching stage, then trains design policies with a singly intractable gradient estimator, enabling cheap multi-policy selection.
-
Input-to-State Stability Certification via Projection Residuals for Koopman Learning Control of Nonlinear Repetitive Systems
A practical input-to-state stability certificate for Koopman learning control is derived, separating prediction residuals from selected-channel margins and projection residuals.
-
Automatic detection of solar filament oscillations I: Multi-scale spectral pipeline
A multi-scale spectral pipeline using deep learning filament detection automatically identifies 91 oscillatory events in two weeks of 2014 GONG data, recovering known events and finding new ones with periods 20-126 min.
-
Causal Inference for Functional Treatments with Stochastic Policies
Develops stochastic policies and single-basis-function modification for causal inference on functional treatments, proves asymptotic normality and rate double robustness, and applies to NHANES physical activity and mortality data.
-
The Representational Limit of Scalar Interactions: An Interventional Decomposition
Signed pairwise interaction scores conflate U/R/S; Stochastic Hi-Fi uses interventional masked inference to recover per-feature uniqueness, redundancy, and synergy profiles.
-
When Representative Samples Produce Worse Outcomes: Scale-up Decisions and Testing in Small-Budget RCTs
In small-budget RCTs where significance tests decide scale-up, optimal pilot sampling shifts from representative to single homogeneous subpopulation as budget shrinks.
-
Expected Free Energy-based Planning as Variational Inference
EFE-based planning is formulated as variational free energy minimization with epistemic priors, decomposing into expected plan costs plus a complexity term.
-
Bayesian Global Fr\'echet Regression via Weak Conditional Expectations
A Bayesian global Fréchet regression method is introduced via a Fréchet Bayes rule that reduces the problem to scalar tasks, allows prior-data interpolation, and remains valid under moment conditions using weak conditional expectations.
-
Statistically and Computationally Optimal Estimation and Inference of Common Subspaces
Establishes statistical and computational optimality thresholds for common subspace estimation and inference under varying SNR regimes, including an impossibility result for adaptive confidence intervals below strong inference SNR.
-
What Type of Inference is Active Inference?
Proper EFE-based planning is VFE plus planning and epistemic entropy corrections, realized by channel-reparameterized message passing that captures novelty.
-
A Van Trees Lower Bound for Fully Interactive Differentially Private Federated Learning
Under clientwise sample-level zCDP, the Fisher information of any fully interactive public federated transcript contracts to a sum of per-client privacy-vs-sample terms, yielding matching minimax rates for mean, linear, and nonparametric regression.
-
Federated Martingale Posterior Samping
Federated martingale posterior sampling lets clients share data embeddings for central predictive Bayesian sampling, matching centralized performance and improving calibration on MNIST, CIFAR-10, and CIFAR-100.
-
Decoupled Conformal Optimisation: Efficient Prediction Sets via Independent Tuning and Calibration
DCO decouples tuning for efficiency from calibration for coverage in conformal prediction, maintaining marginal guarantees and reducing average set sizes on benchmarks like ImageNet-A and Diabetes.
-
Partitioning Neural Co-Variability
The PMNLV model extends single-neuron overdispersion to populations via matrix-normal gain priors, showing shared co-variability highest in V1 and declining along the mouse visual hierarchy.
-
A General Framework for Optimal Group Sequential Testing via Mixed-Integer Linear Programming
The authors propose an S-MILP framework that optimizes group sequential testing boundaries to achieve faster rejection of the null hypothesis compared to traditional methods while controlling type I and type II errors.
-
A Closed-Form Persistence-Landmark Pipeline for Certified Point-Cloud and Graph Classification
PLACE delivers a closed-form certified classification method for point clouds and graphs based on persistent homology with explicit excess-risk bounds, selection rules, and training-time certificates.
-
Deep Variational Inference Symbolic Regression
DVISR performs variational inference over symbolic expression trees and constants by training a neural network with the ELBO as reward, recovering true posteriors in simple test cases.
-
Tweedie Calculus
Conditional expectations of latent variables under additive noise admit unique continuous Tweedie representations via a linear functional constructed from the inverse Fourier transform of an explicit tempered distribution.
-
The Unseen Species Problem Revisited
New theoretical results on estimators and intervals for predicting unseen outcomes in additional samples from discrete distributions, with extensions to grouped incidence data.
-
Variational inference via Gaussian interacting particles in the Bures-Wasserstein geometry
Gaussian particles in a linearized Bures-Wasserstein space perform consensus optimization for variational inference and outperform deterministic gradient methods on low-dimensional non-log-concave targets.
-
Heat-Kernel Entropy Profiles and Geometric Effective Sample Size for Weighted Measures on Manifolds
Heat-kernel smoothing over weighted points on a compact manifold yields a scale-dependent geometric effective sample size that discounts nearby and duplicate particles.
-
Autorelevance function and other feature relevance measures for univariate time series
Introduces autorelevance functions as lag importance measures for time series forecasting using Shapley values and a novel one-step forecast replacement for absent features.
-
A Bayesian latent Gaussian process framework for aerodynamic uncertainty quantification
A Bayesian latent GP calibration framework for aerodynamic surrogates marginalizes input uncertainty and matches output uncertainty statistics, achieving 94.2-95.8% coverage of true 95% intervals.
-
Nonparametric Estimation of Optimal Stochastic Just-In-Time Adaptive Interventions for Distal Outcomes
A nonparametric estimator of the regimen-response curve for stochastic JITAIs on distal outcomes is developed, with weak convergence to a Gaussian process and asymptotic theory for the optimizing policy.
-
Spatio-Temporal Disaggregation with Changing Areal Boundaries
A spatio-temporal disaggregation method that replaces lognormal polygon effects with gamma overdispersion to obtain a marginal negative binomial likelihood, reducing latent variables and enabling fast inference via the Extended Latent Gaussian Model framework.
-
Betting on Moments: Legendre Jumper Martingales for Online Exchangeability Testing
Shifted Legendre polynomials turn Simple Jumper martingales into multi-moment exchangeability tests, with a mean-field variational form that avoids exponential state-space cost.
-
Coupling-Grouped XY-QAOA for Joint Anomaly-Feature Selection
Coupling-Grouped XY-QAOA enables joint anomaly-feature selection via a constraint-preserving grouped-angle QAOA variant, achieving 45.9-61.3% circuit depth reduction and larger feasible executions (64 qubits at p=2) on IBM Heron hardware compared to standard approaches.
-
Optimal sequential two-stage Bayes Factor Design for two-arm clinical Phase II Trials with binary Endpoints
Derives exact operating characteristic corrections and a numerical search over sample sizes to obtain optimal two-stage Bayes factor designs for two-arm binary-endpoint phase II trials that minimize expected sample size under the null.
-
Voronoi-Elitism Genetic Algorithm: A Generic Derivative-Free Routine With Theory and Implementation for Statistical Optimization
VEGA is a derivative-free genetic algorithm that constructs Voronoi neighborhoods around retained elite candidates to balance exploitation and exploration in high-dimensional statistical optimization.
-
Infinite-Dimensional Spherical Kernel ridge Regression
An intrinsic spherical kernel ridge regression framework is introduced for non-linear responses on spheres, reducing infinite-dimensional estimation to finite via the representer theorem with convergence rates shown.
-
Scalable High-Dimensional Bayesian Field Reconstruction with Finite Elements: Application to 3D Porous Media Flow
A finite-element variational inference method delivers full-covariance Bayesian field reconstruction at dimensions exceeding 400,000 for 3D porous media flow using sparse precision parameterization from SPDE priors.
-
Set-Valued Policy Learning
The paper develops set-valued policies and conformal policy learning methods that output treatment sets with marginal coverage guarantees for robust decision-making under uncertainty.
-
Mechanisms of Misgeneralization in Physical Sequence Modeling
Generative sequence models for physical tasks exhibit physical misgeneralization where local prediction errors propagate through physical measurements to distort aggregate distributions over quantities like distance or energy; a data deviation kernel explains and predicts the shifts and supports a内核
-
Pessimistic Risk-Aware Policy Learning in Contextual Bandits
A distributional framework for optimizing Lipschitz risk functionals in offline contextual bandits yields data-dependent suboptimality bounds of Õ(1/√n) that match risk-neutral rates and are minimax optimal.
-
LLMs are not (consistently) Bayesian: Quantifying internal (in)consistencies of LLMs' probabilistic beliefs
LLMs show inconsistent belief updates from evidence, with learned heuristics sometimes beating exact Bayesian computation due to misspecified world models.
-
Information as Maximum-Caliber Deviation: A bridge between Integrated Information Theory and the Free Energy Principle
Information defined as maximum-caliber deviation derives IIT 3.0 cause-effect repertoires from constrained entropy maximization and equates to prediction error under CLT and LDT.
-
LASER: Learning Active Sensing for Continuum Field Reconstruction
LASER couples a latent world model with a GRPO-trained RL policy to adaptively reposition sensors for sparse continuum field reconstruction, outperforming fixed and offline-optimized placement strategies across PDE and real-world benchmarks.
-
Early Academic Capital as the Causal Origin of Dropout in Constrained Educational Systems -- Evidence from Longitudinal Data and Structural Causal Models
Low early academic capital causally raises three-year dropout probability by 25-27 percentage points in a constrained engineering curriculum, roughly twice the effect of later gateway-course repetition.
-
INLA-RF: A Hybrid Modeling Strategy for Spatio-Temporal Environmental Data
A hybrid INLA-RF framework integrates Bayesian spatio-temporal modeling with random forests through two iterative algorithms to improve predictions and uncertainty quantification for environmental data.
-
Compositional amortized inference for large-scale hierarchical Bayesian models
A new error-damping estimator for compositional score matching enables stable amortized inference on hierarchical Bayesian models with over 750,000 parameters using fewer than one full model simulation on large problems.
-
Empirical Calibration and Conditional-Reliability Diagnostics for Bearing RUL Prediction under Operating-Regime Shift
On a 10-bearing PHME subset, a residual-calibrated fusion model hits ~0.15 normalized MAE and 0.90 average 90% coverage under leave-regime-out splits, while conditional diagnostics expose 0.666 coverage and raw-channel-loss collapse.
-
Beyond the Flow: A Bayesian Latent Clustering Framework for Shared Micro-mobility Users in Venice
A Bayesian mixture model with product-multinomial likelihood clusters over 11,000 recurrent users from 220,000+ categorical trips in Venice, yielding eight latent mobility profiles.
-
Incomplete Matrix Regression
IMR models an incomplete matrix as intercepts plus Lasso-penalized covariates plus low-rank latent factors with ridge penalties on known kernels, estimated by modular ALS with non-asymptotic error bounds.
-
Human Capital, AI, and Labor Commoditization
Difference-in-differences analysis around ChatGPT release shows commoditization of labor in AI-exposed job categories on Upwork, with declining human capital importance and rising price importance.
-
Exploring Statistical Change Point Detection Techniques for Performance Anomaly Detection at Mozilla
Ensemble voting strategies for change point detection improve F1-score by 11% over Mozilla's T-test method on a new ground-truth dataset of 174 performance time series annotated by practitioners.
-
A 65 nm Multi-Modal Bayesian Inference Engine with 16.3 fJ/Sample Calibration-Free GRNG for Risk-Aware At-Home Skin Lesion Screening
A 65 nm compute-in-memory chip implements multi-modal Bayesian neural networks with a calibration-free GRNG to deliver risk-aware skin lesion screening with reported gains in coverage, robustness, and efficiency over unimodal baselines.
-
DXA-Derived Skeletal Phenotypes and Hip Fracture Risk: A Backdoor-Adjusted Causal Analysis
Backdoor-adjusted ATEs on 21,098 UK Biobank participants showed total femur BMC and BMD with the largest hip fracture risk reductions (-0.0047 per SD), and adding the top 11 phenotypes to clinical variables raised AUC to 0.842 versus FRAX 0.709.
-
The unique, universal entropy for complex systems
Coupled entropy maximized by coupled stretched exponentials is the unique universal entropy meeting scale-specific uncertainty measurement and Hanel-Thurner extensivity requirements.