Papers

Batch normalization (BN) is central to modern deep networks, but its effect on the realized function during training remains less understood than its optimization benefits. We study training-time BN in continuous piecewise-affine (CPA) networks through the geometry of switching hyperplanes and the induced affine-region partition. Conditioned on a mini-batch, we show that BN defines for each neuron a reference hyperplane through the batch centroid, and that breakpoint-switching hyperplanes are parallel translates whose offsets are expressed in batch-standardized coordinates and are independent of the raw bias. This yields an exact criterion for when a switching hyperplane intersects a local $\ell_\infty$ window and motivates a local region-density functional based on exact affine-region counts. Under explicit sufficient conditions, we show that BN increases expected local partition refinement in ReLU and more general piecewise-affine networks, and that this mechanism transfers locally through depth inside parent affine regions where the upstream representation map is an affine embedding. These results provide a function-level geometric account of training-time BN as a batch-conditional recentering mechanism near the data.

We study long-horizon deployment of a frozen predictor under dynamic covariate shift. A time-domain Poincare inequality first reduces temporal risk volatility to derivative energy. A Jacobian-velocity theorem then supplies the corresponding pathwise control. Given explicit regularity and domination assumptions, the theorem identifies directional tangent energy along the deployment path as the governing quantity. Under low-rank drift, that quantity reduces to directional Jacobian energy in the drift subspace, motivating drift-aligned tangent regularization (DTR) and a matched monitoring proxy. Rather than smoothing the network isotropically, DTR penalizes sensitivity only along estimated drift directions. We validate the theorem-to-method pipeline in four experiments: a synthetic benchmark for the time-domain inequality, a controlled synthetic comparison against isotropic Jacobian regularization, and two frozen-deployment studies on the UCI Air Quality and Tetouan power-consumption datasets. DTR reduces risk volatility and directional gain in the controlled low-rank regime and beats isotropic smoothing there. It also gives validation-selected deployment gains on both real datasets, with the Air Quality subspace estimated from target-orthogonal sensor motion. Moderate drift-subspace misspecification is tolerable while orthogonal misspecification largely removes the benefit.

Despite theoretical advantages, causal methods for Gene Regulatory Network (GRN) inference from single-cell RNA-seq data consistently fail to match or outperform correlation-based baselines in many realistic benchmarks, a persistent puzzle which casts doubt on the value of causality for this task. We argue that existing benchmarks are insufficiently controlled to answer this question because they evaluate on real or semi-real data where multiple pathologies co-occur, confounding failure modes, and obscuring the specific conditions under which different inference methods excel or fail. To address this gap, we introduce a controlled diagnostic framework that isolates seven biologically motivated pathologies (dropout, latent confounders, cell-type mixing, feedback loops, network density, sample size, and pseudotime drift) and measure how six representative methods spanning three inference paradigms degrade as each pathology intensifies. Across 6,120 controlled experiments, we find that causal methods genuinely dominate in clean and structurally favorable regimes, but specific pathologies (notably dropout and latent confounders) selectively neutralize their advantages. We further introduce an error-type decomposition that reveals methods with similar aggregate accuracy commit qualitatively different errors. To probe whether single-pathology effects persist when multiple stressors co-occur, we perform an interaction sweep over the three most impactful pathologies and find that their joint effects are sub-additive, while also exposing density-conditional cross-overs invisible to single-dial analysis. Our findings offer a nuanced understanding of when and why different methods succeed or fail for GRN inference, providing actionable insights for method development and practical guidance for practitioners.

Sea surface temperature (SST) is a fundamental determinant of global climate dynamics and economic activity. Reliable projections of future SST patterns depend critically on a rigorous characterization of the underlying spatial random field. In this study, we introduce a novel convolution-based covariance framework tailored to geostatistical domains constrained by physical barriers and influenced by vector-driven flows. By discretizing the continuous marine domain into a directed linear network that preserves the orientation of ocean currents, we construct a moving-average stochastic process whose dynamic is encoded via a Markovian transition-probability matrix on the network's vertices. The induced covariance structure emerges as a weighted combination of a spatial kernel and flow-dependent weights, giving rise to a complex estimation problem. To stabilize inference, we propose a penalized estimator that regularizes covariance parameters while enforcing consistency with known hydrodynamic properties. We then embed this covariance model into a Monte Carlo simulation framework to refine RCP-based SST projections and to identify thermal 'hot spots' of heightened ecological risk. Our approach delivers a statistically principled framework that prevents physical inconsistencies – such as correlations across land barriers – providing a robust basis for quantifying uncertainty in future SST forecasts and for guiding targeted environmental assessments.

We study the composite sequential quantum hypothesis testing (SQHT) problem, where the objective is to distinguish a null quantum state from a set of alternative quantum states. We propose a mixture-sequential quantum probability ratio test that adaptively selects measurements based on the current mixture estimate of the alternative set, and stops upon the first threshold crossing of the mixture log-likelihood ratio. Under an expected sample size constraint, we show that our proposed strategy simultaneously achieves the Type-I and (worst-case) Type-II error exponents, characterized by the minimal measured relative entropies between the null state and the alternative set. We further establish a matching converse, thereby characterizing the optimal error exponent region. Finally, our results show that achieving vanishing error probabilities in composite SQHT requires an expected sample complexity at least as large as that of sequential testing between two fixed states.

The duality $L^{\infty}\simeq (L^{1})'$ frequently breaks down in the presence of model uncertainty, where a single reference measure $P$ is replaced by a non-dominated family of probability measures $\mathcal{P}$. The unavailability of classical measure-theoretic and functional-analytic tools in this regime poses a significant obstacle to developing robust probabilistic frameworks. We show that this duality can be restored for a broad class of robust statistical models by extending the underlying probability space. Specifically, on the extended model, the space $\mathbb{L}^{\infty}(\mathcal{P})$ of $\mathcal{P}$-quasi-surely bounded functions is isometrically isomorphic to the dual of the space of finite signed measures absolutely continuous with respect to at least one element of $\mathcal{P}$. The proposed extension is canonical: it is the smallest $\mathcal{P}$-complete extension of the original $\sigma$-algebra for which $\mathbb{L}^{\infty}(\mathcal{P})$ is the dual of any normed space. Our assumptions encompass several prominent non-dominated settings, including infinite product measures, Gaussian processes, the Black-Scholes model with uncertain constant volatility and drift, robust binomial models, and, more generally, infinite sequences from any parametric model with almost surely estimable parameters. Furthermore, we unify the existing frameworks of Cohen (2012) and Liebrich et al. (2022), demonstrating that our construction is equivalent to the capacity-based approach under mild assumptions satisfied by the aforementioned examples. Finally, we apply our theory to extend Kraft's (1955) characterization of strictly unbiased hypothesis tests to non-dominated cases.

Published transfer-BO comparisons often estimate an average treatment effect of acquisition choice over hidden regime variables, while practitioners need the conditional effect for their specific prior quality, budget ratio, and metric. An audit of 40 transfer-BO papers from NeurIPS, ICML, ICLR, AISTATS, UAI, TMLR, JMLR, and AutoML-Conf (2022-2025) finds that 98% never vary B/|A| as a controlled axis. On the same GDSC2 benchmark, changing only the budget reverses the ranking: at B=50, Greedy outperforms UCB by 0.050 Hit@1, while at B=100, UCB outperforms Greedy by 0.035. We capture this transition with the Portable Regime Score PRS=(B/|A|)(1-rho), where rho is the prior rank correlation and can be estimated from pilot contexts before the main comparison. Across 79 conditions spanning chemistry, drug-response biology, and HPO, a hierarchical model gives beta=0.50 (p=1.1e-9), and 19% of conditions fall in an equivalence zone where |advantage|<0.01 Hit@1. In five published reversal cases, PRS predicts the winner from pre-comparison observables. A No-Free-Leaderboard proposition explains why unconditional rankings are unstable: when CATE changes sign across regimes, the reported ATE becomes a function of benchmark mixture. RegimePlanner, which estimates rho online and switches acquisition accordingly, wins all 16 HPO-B search spaces at B=100 and exceeds the matched {Greedy,UCB} per-context oracle on GDSC2 by 18%. Pre-registered predictions achieve 27/40=67.5% overall accuracy and above 90% within EMA prior families. The practical protocol is simple: report B/|A|, rho, K, and metric alongside any claimed acquisition advantage.

Every attention head defines a degree-normalized transport operator, and a growing family of diagnostics reads model behavior (hallucination among them) from its spectrum. We ask what such diagnostics can and cannot infer. The operator splits orthogonally into a symmetric part governing transport capacity and an antisymmetric part encoding orientation. We prove an identifiability limit: every transpose-invariant spectral diagnostic is orientation-blind (unable to distinguish an operator from its transpose, hence blind to the orientation of information flow), with a transpose-stability bound limiting any Lipschitz diagnostic's transpose sensitivity by the asymmetry coefficient $G$. This bounds what spectral diagnostics of the attention operator can resolve (e.g.\ LapEigvals and the attention-spectral branch of LLM-Check). On the surviving axis, a closed-form bipartite-Cheeger landscape shows uniform causal attention obeys an $n$-independent temporal-cut floor $\phi \ge 1/5$ while window attention pierces it as $O(w/n)$; the floor is an idealized benchmark, not an empirical attractor, and the fraction of real heads falling below it is itself an empirically stable architectural descriptor. The two-axis diagnostic ($\phi$ for capacity, $G$ for asymmetry magnitude) yields a falsifiable polarity prediction, borne out in sign under length-controlled, forced-scoring evaluation across decoder-only, encoder-only, and encoder–decoder models (capacity-axis signal 0.62–0.84 LC-AUROC): polarity reverses between HaluEval and MedHallu, directionally as predicted though asymmetric in strength, with decision polarity calibrated per regime.

The standard constraint-based paradigm for causal discovery with incomplete data – impute first, test second – is frequently miscalibrated: any consistent conditional independence (CI) test rejects a true null with probability approaching 1 when imputation error induces spurious conditional dependence. We introduce PAIR-CI, a nonparametric CI test that restores calibration by integrating multiple imputation directly into the inferential procedure via a paired permutation design. PAIR-CI compares cross-validated models that include and exclude the candidate variable while receiving the same imputed conditioning set, forcing imputation error to cancel in their loss difference rather than contaminate the test statistic. A provably consistent variance estimator jointly accounts for uncertainty arising from cross-validation and multiple imputation – to our knowledge, the first formal unification of these two inferential frameworks. In simulations, existing imputation-based CI tests exhibit false positive rates of 28–45% when data are missing not at random (MNAR), whereas PAIR-CI averages below the nominal 5% level across data-generating processes and missingness mechanisms. These gains are largest in nonlinear settings and grow with causal graph size: when integrated into the PC algorithm, PAIR-CI reduces structural Hamming distance by 8% on 10-variable nonlinear graphs, 15% on 30-variable equivalents, and up to 44% on the 56-variable HAILFINDER network, with stable performance in all settings.

Protecting confidential data while preserving utility is particularly challenging when data sets contain outlying observations. Existing latent space anonymization methods, such as spectral anonymization (SA), rely on principal component analysis (PCA) and may therefore be vulnerable to contamination. We investigate anonymization in the presence of outliers and propose ICSA, a robust alternative to SA based on invariant coordinate selection (ICS). By replacing the PCA transformation with ICS, the robustness of the anonymization procedure can be regulated through the choice of scatter matrices. Alongside the methodological development, we derive a theoretical result showing that SA fails under sufficiently influential outliers. To assess the practical implications of this result, we compare the privacy-utility trade-off of ICSA and SA through simulation experiments under varying contamination settings and outlier severities. Our findings indicate that implementations of ICSA based on robust scatter matrices achieve stronger privacy protection than SA, while typically maintaining comparable, and in some cases improved, utility. We further examine the empirical performance of the proposed method using a benchmark clinical data set, where ICSA demonstrates superior overall privacy-utility efficiency relative to SA. These results suggest that explicitly accounting for outliers can materially improve anonymization performance and that robust latent space transformations offer a promising direction for privacy-preserving statistical data release.

The paper applies a new, nonparametric, interpoint distance-based measure to confirm the inherent groups prevailing in the brightest source of light in the universe: gamma-ray bursts. Our effective metric, in association with clustering methods like Gaussian-mixture model-based and $K$-means algorithms, resolves the conflict regarding the possibility about existence of more than binary clusters in the gamma-ray burst population. Here we carry out multiple nonparametric statistical tests of hypotheses, as many as the number of bursts available from the `BATSE' catalog. An integrated $p$-value achieved from the aforesaid dependent tests solves our concern confirming two groups of short and long bursts.

Cross-frequency interactions are fundamental brain mechanisms for integrating information across temporal scales. However, accurate identification of these couplings is hindered by complex multi-frequency nonlinearities and by spurious, zero-lag artifacts caused by volume conduction. To our knowledge, conventional metrics lack a robust framework to characterize genuine interactions among multiple time series where a frequency of interest $f_N$ arises from the combination of $N-1$ components such that $f_N = \sum_{i=1}^{N-1} f_i$. We introduce a general family of antisymmetric cross-polyspectral indices designed to quantify these harmonic dependencies while being intrinsically robust to instantaneous mixing. We derive the theoretical properties of these quantities and validate them through simulations of cubic nonlinearities. As a proof of concept, we apply the indices to empirical EEG recordings; the results reveal significant higher-order dependencies that elude standard analytical approaches. We further discuss how these indices can inform novel, personalized multi-site transcranial magnetic stimulation (mTMS) protocols by enabling the selective monitoring and modulation of specific multi-frequency network interactions.

Understanding a dynamic network requires more than locating when it changes: we also want to know the magnitude and structural direction of change and which nodes contribute. Unfolded spectral embedding places network snapshots in a joint representation, but the underlying rectangular latent factorization is invariant under general linear transformations. Such transformations preserve edge probabilities while potentially distorting Euclidean second moments. We introduce Multiscale Euclidean Network Trajectories (MENT), whose central estimands are pairwise second-moment operators of the dynamic latent positions. Under an isotropic normalization of the shared anchor latent positions, these operators are identifiable up to simultaneous orthogonal conjugation. They yield a global trace variation distance and, through an aggregated operator, shared orthogonal directions for mode-wise distances. The squared global distance decomposes exactly across these modes. Classical multidimensional scaling turns the distances into global and mode-wise trajectories of time. A modified unfolded spectral estimator consistently recovers the operators and induced trajectories, and we establish bounds connecting node attributions to trajectory displacements and trajectory estimation to change point localization. Synthetic experiments validate geometric recovery and show improved detection of mode-specific changes against baselines. Analyses of two real networks illustrate interpretable mode- and node-level descriptions of temporal evolution.

The distributional statistical framework replaces classical probability densities by distribution-kernel pairs $(T, \varphi)$, where $T$ is a tempered distribution and $\varphi$ is a rapidly decaying kernel. We develop the thesis that the kernel acts as a geometric regulariser, placing parametric statistical models in generic (transversal) position relative to degeneracy loci encoding non-identifiability, singular information, moment indeterminacy, and representation failure. Using the transversality theorems of Whitney, Thom, and Mather, we prove a finite-dimensional weak transversality theorem: for a generic kernel in any sufficiently rich family, the kernel-induced feature map avoids degeneracy strata of sufficiently high codimension. We establish verifiable conditions – formulated as rank conditions on the Jacobian of the joint feature map – under which the transversality hypothesis can be checked, and verify them for location families, the log-normal, Stein discrepancies, and graphical models. The present results apply to parametric models; extensions to semiparametric and nonparametric settings are discussed. The degeneracy classification includes representation degeneracy (Type 0) for models without closed-form densities and higher-order instabilities (Type IV) in non-chordal graphical models. Identifiability, robustness, moment determinacy, Fisher information regularity, Stein discrepancy, inferential separation, and the Behrens-Fisher problem all admit a unified geometric interpretation as transversality conditions on the feature map. This paper serves as a geometric companion to a series of papers developing the distributional framework.

Inferring continuum models directly from video is hampered by two facts: the recorded field is uncalibrated image intensity rather than a physical state, and direct numerical differentiation of noisy frames is unstable. We develop a video-to-PDE pipeline that converts grayscale recordings of an ink plume into a normalised scalar field $u(x,y,t)$, isolates a bulk drift $\mathbf{v}(t)$ from intrinsic spreading via the intensity-weighted centroid, and identifies an effective transport law by weak-form sparse regression. Conditioning, threshold-sweep and random-centre diagnostics show that overcomplete libraries are strongly collinear; the search is therefore restricted to compact gradient-based libraries. Coefficients are refined by an inverse physics-informed network and recalibrated against forward rollouts, with a chronological block bootstrap quantifying uncertainty. The selected reduced model $u_t+\mathbf v(t)\!\cdot\!\nabla u = 9.005\,|\nabla u|^{2}+0.666\,\Delta u$ outperforms advection–diffusion baselines on held-out frames, retains a positive Laplacian coefficient, and admits a Cole–Hopf reduction to a linear advection–diffusion equation. The framework demonstrates that uncalibrated visual data can yield compact, predictive and structurally interpretable continuum models when discovery, calibration and uncertainty are treated as distinct stages.

Single-cell ATAC-seq (scATAC-seq) enables high-resolution mapping of chromatin accessibility, yet privacy regulations and data size constraints hinder multi-institutional sharing. Federated learning (FL) offers a privacy-preserving alternative, but faces three fundamental barriers in scATAC-seq analysis: ultra-high dimensionality, extreme sparsity, and severe cross-institutional heterogeneity. We propose FL-Sailer, the first FL framework designed for scATAC-seq data. FL-Sailer integrates two key innovations: (i) adaptive leverage score sampling, which selects biologically interpretable features while reducing dimensionality by 80%, and (ii) an invariant VAE architecture, which disentangles biological signals from technical confounders via mutual information minimization. We provide a convergence guarantee, showing that FL-Sailer converges to an approximate solution of the original high-dimensional problem with bounded error. Extensive experiments on synthetic and real epigenomic datasets demonstrate that FL-Sailer not only enables previously infeasible multi-institutional collaborations but also surpasses centralized methods by leveraging adaptive sampling as an implicit regularizer to suppress technical noise. Our work establishes that federated learning, when tailored to domain-specific challenges, can become a superior paradigm for collaborative epigenomic research.

Large-scale population-level datasets, such as the UK Biobank and the All of Us Research Program, often lack covariates needed for a specific analysis, such as genetic or lifestyle measures, while related studies measure them. This creates a cross-population missing data problem in which covariates are completely unobserved in the target population, rather than partially missing within one dataset. We propose an augmented transfer regression learning method for this setting. The key identifying condition is a sub-population shift assumption: the joint distribution of the outcome and observed covariates may differ across source and target populations, but the conditional distribution of the missing covariates given observed variables is invariant. We combine importance-weighted estimating equations with imputation terms for first- and second-order moments of the missing covariates. The resulting estimator is doubly robust and remains consistent if either the density ratio model or both imputation models are correctly specified. It is n^{1/2}-consistent and asymptotically normal, and attains the semiparametric efficiency bound when both

Model selection plays an important role in longitudinal data analysis, especially when models are estimated using the generalized method of moments (GMM) in the presence of time-dependent covariates. In this setting, the number of valid moment conditions can grow quickly and may lead to over-parameterized models. The Kullback–Leibler Information Criterion (KLIC) has been proposed as a model-selection tool for this framework; however, the original KLIC criterion may favor overly complex models when the number of parameters or valid moment conditions increases. To address this limitation, this study proposes two penalized versions of KLIC that incorporate penalties based on both the number of model parameters and the number of valid moment conditions. The proposed criteria are referred to as the Moment–Parameter Product Penalty KLIC (MPPP–KLIC) and the Logarithmic Penalty KLIC (LP–KLIC). These criteria provide a theoretically motivated mechanism for balancing model fit and model complexity in GMM-based longitudinal models. Through an extensive simulation study involving both binary and continuous response settings, the proposed criteria are shown to improve the ability of KLIC to distinguish among competing models and to reduce the selection of over-parameterized models. The performance of the proposed methods is further illustrated using the Filipino Child Morbidity dataset, a longitudinal study of child health in the Philippines. The results show that the proposed penalized criteria provide stable and interpretable model rankings and consistently identify age as the most important predictor of child morbidity. Overall, the proposed penalized KLIC criteria offer practical and theoretically grounded tools for model selection in GMM-based longitudinal data analysis with time-dependent covariates.

High-dimensional neuroimaging and spatiotemporal blocks often contain structured missingness from acquisition artifacts, preprocessing failures, and sensor dropout. Multiple imputation propagates uncertainty, but fully conditional specification methods such as multivariate imputation by chained equations (MICE) can be slow or unstable when block dimension is large and correlations are strong. A multivariate normal (MVN) working model provides a coherent posterior predictive target and an exact data augmentation sampler, but repeated covariance sampling and matrix factorizations become costly in large dimensions. We propose High-dimensional Imputation via covariance Mode and Chained Equations (HIMCE), a hybrid multiple-imputation procedure for continuous blocks. Relative to exact MVN data augmentation, HIMCE preserves the Gaussian conditional imputation law and propagates mean- parameter uncertainty through stochastic coefficient or local-ridge draws. In high-dimensional blocks, it approximates covariance uncertainty through covariance-mode updating, optionally with a scalar bridge; in small blocks, it can restore exact covariance uncertainty through a conditional inverse-Wishart refresh. We record the exact Bayesian reference sampler and prove fixed-dimensional posterior consistency and asymptotic equivalence of mode plug-in prediction in total variation. We also develop diagnostics based on randomized rank-cell probability integral transform (PIT), PIT-consistent empirical coverage, and marginal distribution overlays. In the primary spatial benchmark, HIMCE improves posterior-mean error relative to HIMA and screened MICE, runs at HIMA-like speed and below half the MICE runtime, and improves interval coverage over HIMA, although MICE remains better calibrated. A repeated low- dimensional NHANES illustration shows improved coverage with competitive point prediction.

Structural causal models provide a unified semantics for interventions and counterfactuals, but most identifiability results rely on restrictive assumptions like global monotonicity, which are often violated in embodied interaction, where the same exogenous perturbation can induce opposite responses under different contact contexts. We ask what structure still suffices once global monotonicity is dropped. We introduce non-monotone triangular structural causal models (NM-TM-SCM), which retain triangular recursion but replace global monotonicity with mechanism-wise invertibility and context-independent inverse transport. We prove that these conditions are equivalent to exogenous isomorphism and imply complete counterfactual identifiability, and we give a counterexample showing that local invertibility alone is insufficient. We instantiate the theory in CausalInverter, with triangular invertible layers, orientation gates, and transport-stability regularization. On synthetic non-monotonic mechanisms, the structural bias yields systematic counterfactual gains as non-monotonicity increases. On MuJoCo Door, our model achieves perfect event-level counterfactual recovery, lowers continuous angle error relative to a Transformer baseline, and delivers substantially more stable recovery than Transformer and conditional-flow predictors. On MuJoCo Push, where non-monotonicity is weaker, the same low-data predictors remain competitive or better, consistent with a bias-variance boundary. These results identify a broader identifiable regime between globally monotone triangular models and unconstrained black-box world models.

Causal discovery methods such as LiNGAM identify causal structure from observational data by assuming mutually independent disturbances. This assumption is fragile: shared volatility, common scale effects, or other forms of dependence can cause the methods to recover the wrong causal order, even with infinite data. We introduce the Linear Mean-Independent Acyclic Model (LiMIAM), which replaces full independence with weaker one-sided mean-independence restrictions on the disturbances. Under finite-order consequences of these restrictions, source nodes are generically identifiable, and hence a compatible causal order can be recovered recursively. Our proof is constructive and leads to DirectLiMIAM, a sequential residual-based algorithm for causal discovery under dependent noise. In simulations with mean-independent but dependent disturbances, DirectLiMIAM outperforms LiNGAM methods. A large-scale empirical application to the oil market highlights the implausibility of the independence assumption and the ability of DirectLiMIAM to recover a realistic causal ordering, from policy to production and from prices to inflation.

The role of the microbiome in disease pathogenesis is an emerging field with strong evidence suggesting that dysbiosis is associated with precancerous and cancerous states. Microbiome data present substantial challenges for causal mediation analysis due to sparsity, compositional constraints, and latent heterogeneity. To address these issues, we propose a zero-inflated beta mixture (ZIBM) method for mediation analysis with compositional microbiome mediators. The proposed method accommodates excess zeros through a zero-inflation component and captures heterogeneity in non-zero relative abundances using a beta mixture distribution. Within the potential-outcomes framework, the ZIBM provides estimates of marginal microbiome-mediated causal effects, and model parameters are estimated using an expectation-maximization algorithm. Simulation studies demonstrate that the ZIBM yields more accurate estimation and reliable inference under conditions commonly observed in microbiome data, compared with existing approaches. An application to a real microbiome study further illustrates its practical utility. These results indicate that the proposed method provides a more flexible and robust statistical framework for mediation analysis involving compositional microbiome data.

Climate-driven outages pose a growing threat to cyber-physical power system (CPPS) resilience, particularly as modern grids increasingly rely on communication, sensing, and control infrastructure for situational awareness and coordinated response. Empirical outage studies and interdependency-aware resilience analyses are often studied separately, limiting our understanding of how observed climate-risk patterns translate into cascading degradation across coupled cyber and physical layers. This paper presents a data-calibrated stress-testing framework for joint power-communication networks. Using EAGLE-I outage records from 2015-2023, we characterize national climate-related outage patterns and identify severe-risk contexts defined by outage duration and customer impact and associated with event type, season, and geography. EAGLE-I customer-impact percentiles are log-normalized and scaled into representative stress intensities for spatially clustered benchmark scenarios on an IEEE 118-bus power system with an overlaid communication network. The Modified Implicative Interdependency Model (MIIM) is used to simulate cross-layer cascade propagation and quantify post-event operability, resilience gaps, and affected cyber-physical entities. Results show that even the highest-severity benchmark scenario degrades post-cascade operability to approximately 60%, producing a resilience gap roughly twice that of the inland baseline. The findings suggest that resilience assessment based only on outage statistics may miss bounded but meaningful amplification effects in interdependent power-communication infrastructure. The proposed framework provides an initial step toward data-calibrated, interdependency-aware resilience assessment for extreme-weather-affected CPPS.

Droughts and flash droughts (rapidly developing droughts; FDs) remain impactful events that are known to desiccate landscape and destroy crops. In particular, droughts in Africa are often more impactful than in other locations, such as the United States or Europe, due to many regions in Africa heavily depending on local agriculture for sustenance. In recent years, large machine learning (ML) models, such as GraphCast and AIFS, have emerged as effective tools for global weather prediction. However, sparse data observations and few ML studies in Africa have left it unclear if these ML models retain their skill when focused on Africa. As such, this project seeks to examine the predictability of drought and FD in Africa using a CrossFormer model based on the Community Research Earth Digital Intelligence Twin (CREDIT) framework developed by NSF NCAR. Our CrossFormer model, termed DroughtFormer, incorporates variables from the ERA5 and GLDAS2 reanalyses and the IMERG and MODIS satellite observations, and employs dry air mass and moisture conservation, to predict soil moisture, vegetation health, and other drought-related surface variables. While DroughtFormer displayed lower accuracy in predicting precipitation and FD indices, it showed significant skill in predicting the remaining variables, delivering stable and skillful forecasts out to 90-day lead times (either beating out or having comparable skill to climatology). In particular, DroughtFormer skillfully represented climate anomalies for key variables, such as soil moisture (though it struggled with the magnitude of the anomalies). Thus, DroughtFormer showed significant promise in representing and predicting agricultural level drought in a region that is heavily impacted by drought events.

This paper develops a statistical theory of extrapolation for large language models, by reinterpreting them through pre-post-additive noise models. In contrast to the standard autoregressive next-token prediction based on an exact prefix, we introduce a perturbation-based procedure that first transforms the prefix into a semantic neighbour and then conditions on this perturbed variant for next-token prediction. This yields a hierarchical model with a pre-post-additive noise structure. Within this framework, we develop a rigorous theory of extrapolability, namely, the capacity of a model class to make reliable predictions for token sequences that lie outside the empirical support of the training corpus, by establishing five properties of the proposed procedure: adaptivity, contractivity, robustness, extrapolability, and double robustness. We evaluate the finite sample performance of the proposed procedure using both synthetic and real world language data. Results show that the proposed method consistently improves out-of-support prediction while maintaining competitive in-support performance, demonstrating that perturbation offers a practical route to language modelling.

Reliable adaptive beamforming is critical for large microphone arrays operating in highly dynamic acoustic environments. In scenarios characterized by fast-moving talkers and interferers, the available sample support for estimating the spatial correlation matrix is often snapshot-deficient. This deficiency, coupled with array imperfections, degrades the White Noise Gain (WNG), leading to severe target signal cancellation. To ensure stable and robust beamforming, we propose a novel adaptive diagonal loading method that guarantees the WNG remains strictly within specified bounds. By leveraging the Kantorovich inequality, we map the desired WNG to a strict upper bound on the condition number of the correlation matrix. Furthermore, we present three estimation techniques for the adaptive loading level, ranging from trace-based bounding to exact eigenvalue decomposition, offering scalable computational complexities of $\mathcal{O}(M)$, $\mathcal{O}(M^2)$, and $\mathcal{O}(M^3)$. Our approach demonstrates highly stable beamforming under fast-changing interference.

Identifying governing equations for a dynamical system is a topic of critical interest across an array of disciplines, from mathematics to engineering to biology. Machine learning – specifically deep learning – techniques have shown their capabilities in approximating dynamics from data, but a shortcoming of traditional deep learning is that there is little insight into the underlying mapping beyond its numerical output for a given input. This limits their utility in analysis beyond simple prediction. Simultaneously, a number of strategies exist which identify models based on a fixed dictionary of basis functions, but most either require some intuition or insight about the system, or are susceptible to overfitting or a lack of parsimony. Here we present a novel approach that combines the flexibility and accuracy of deep learning approaches with the utility of symbolic solutions: a deep neural network that generates a symbolic expression for the governing equations. We first describe the architecture for our model, then show the accuracy of our algorithm across a range of classical dynamical systems.

We introduce a novel approach to finite sample robustness that avoids the pessimism of traditional breakdown analyses. We define the threshold breakdown point, the smallest contamination fraction needed to induce a prescribed deviation, and the finite sample m-sensitivity, the worst-case deviation that an estimator can incur after m observations are contaminated. We derive these measures for commonly used M-estimators, their standard errors and related test statistics. This allows us to extend the decision breakdown point of Zhang (1996) to obtain general breakdown characterizations for hypothesis testing, and show how these notions correspond to finite sample counterparts of the power and level breakdown functions of He, Simpson and Portnoy (1990). We complement our work with an inferential framework for the threshold breakdown and m-sensitivity that yields consistency and asymptotic normality results, as well as a valid multiplier bootstrap for uncertainty quantification. We illustrate the practical utility of our methods in various numerical examples and an application to a two sample testing problem for a blood pressure dataset.

Quantile shares, introduced by Babichenko, Feldman, Holzman, and Narayan [STOC 2024], offer an ordinal, self-maximizing, and interpretable benchmark for fair division of indivisible goods, but their universal feasibility is known only conditional on the rainbow Erdős matching conjecture (EMC). Specifically, Babichenko et al. showed that assuming the rainbow EMC in the near-perfect matching regime, the $(1/2e)$-quantile share is universally feasible. In contrast, a simple argument shows that the $q$-quantile share can be infeasible for any $q > 1/e$. We introduce a one-parameter refinement of quantile shares, the $c$-thinned quantile share, obtained by thinning the inclusion probability in the random benchmark bundle by a factor of $c$ for a fixed constant $c\in(0,1]$. Our main result is that there exists a universal constant $c >0$ for which the $c$-thinned $e^{-c}$-quantile share is unconditionally universally feasible; this is best possible in the sense that for any $c \in (0,1]$, the $c$-thinned $q$-quantile share can be infeasible for any $q > e^{-c}$. Prior to this work, the only nontrivial share known to be universally feasible was Feige's residual maximin share. The thinning viewpoint also lets us remove the factor-two loss in the conditional result for the original quantile share: assuming the rainbow EMC, the $(1/e)$-quantile share is universally feasible.

Climate classification plays a vital role in agricultural planning, hydrological studies, and climate science. One of the most widely used systems for classifying global climate zones is the Köppen-Trewartha (KT) classification. However, the KT classification is fundamentally deterministic, offering discrete labels to spatial locations without accounting for uncertainties in classification. In this paper, we provide a framework for probabilistic modeling of climatic zones. We implement a feedforward artificial neural network (ANN) for classification, allowing for efficient, uncertainty-aware categorization of climatic regions, thereby offering a more nuanced understanding of transitional climate zones compared to traditional deterministic methods. We apply this method to the Sahara Desert region over the 30-year period of 1960 - 1989, using data at more than 400,000 space-time locations from the first 11 years to train our model. We assess the model's short- and long-term classification capabilities to evaluate its stability and accuracy over time. We also compare the probabilistic classification from our model with the traditional KT classification. In addition, we use fluctuation analysis methods to highlight the temporal evolution of climatic zones across the Sahara region and identify areas undergoing significant flux of probabilities of their climate classes, providing insights into broader trends in desertification.