arXiv subjects

Gesine Reinert

Publications and source records attributed to Gesine Reinert.

At least 19 recordsLinked to original sources

arXiv2026

MSGNN: A Spectral Graph Neural Network Based on a Novel Magnetic Signed Laplacian

Signed and directed networks are ubiquitous in real-world applications. However, there has been relatively little work proposing spectral graph neural networks (GNNs) for such networks. Here we introduce a signed directed Laplacian matrix, which we call the magnetic signed Laplacian, as a natural generalization of both the signed Laplacian on signed graphs and the magnetic Laplacian on directed graphs. We then use this matrix to construct a novel efficient spectral GNN architecture and conduct extensive experiments on both node clustering and link prediction tasks. In these experiments, we consider tasks related to signed information, tasks related to directional information, and tasks related to both signed and directional information. We demonstrate that our proposed spectral GNN is effective for incorporating both signed and directional information, and attains leading performance on a wide range of data sets. Additionally, we provide a novel synthetic network model, which we refer to as the Signed Directed Stochastic Block Model, and a number of novel real-world data sets based on lead-lag relationships in financial time series.

stat.ML↗

arXiv2026

Stein's method in network analysis

The paper consists of a brief survey of the use of Stein's method in network analysis. Topics covered include normal and Poisson approximation of graph--based statistics, approximating an exponential random graph by a Bernoulli model, and comparison of different random geometric graph models.

math.PR↗

arXiv2026

TASER: Task-Aware Stein Regularisation for Geometry-Driven Robustness

Modern deep networks remain fragile under distribution shift and adversarial perturbations, often due to excessive or poorly structured input sensitivity. We introduce TASER (Task-Aware Stein Regularisation), a training-time regularisation framework derived from Langevin Stein operators. By penalising pointwise Stein residuals under the training distribution, TASER encourages geometric compatibility between predictors and data density, inducing anisotropic, data-aware smoothness. We provide theoretical links between Stein regularisation and reduced first-order shift sensitivity, develop scalable implementation variants compatible with modern architectures, and demonstrate improved robustness and stability across regression and vision benchmarks. Across CIFAR-10 experiments, TASER consistently improves the adversarial robustness of established training methods without incurring statistically significant clean-accuracy degradation.

cs.LG↗

arXiv2026

A Pure Hypothesis Test for Inhomogeneous Random Graph Models Based on a Kernelised Stein Discrepancy

Complex data are often represented as a graph, which in turn can often be viewed as a realisation of a random graph, such as an inhomogeneous random graph model (IRG). For general fast goodness-of-fit tests in high dimensions, kernelised Stein discrepancy (KSD) tests are a powerful tool. Here, we develop a KSD-type test for IRG models that can be carried out with a single observation of the network. The test applies to a network of any size, but is particularly interesting for small networks for which asymptotic tests are not warranted. We also provide theoretical guarantees.

stat.ML↗

arXiv2026

Stein's method of moment estimators for local dependency exponential random graph models

Providing theoretical guarantees for parameter estimation in exponential random graph models is a largely open problem. While maximum likelihood estimation has theoretical guarantees in principle, verifying the assumptions for these guarantees to hold can be very difficult. Moreover, in complex networks, numerical maximum likelihood estimation is computer-intensive and may not converge in reasonable time. To ameliorate this issue, local dependency exponential random graph models have been introduced, which assume that the network consists of many independent exponential random graphs. In this setting, progress towards maximum likelihood estimation has been made. However the estimation is still computer-intensive. Instead, we propose to use so-called Stein estimators: we use the Stein characterizations to obtain new estimators for local dependency exponential random graph models.

math.ST↗

arXiv2026

Approximating inter-point distances in directed Bernoulli graphs

In directed random graphs, in which edges can be assigned to have one of two directions, or perhaps both, the distance between two vertices $v$ and $v'$ can be computed along paths that are directed from $v$ to $v'$, or along paths that are directed from $v'$ to $v$. These two distances are in general dependent. Here, we approximate their joint distribution in the setting of the directed Bernoulli random graph $\mathcal{DG}(n,p,θ)$, obtained as a natural extension of the Bernoulli random graph $\mathcal{G}(n,p)$ by assigning directions to the edges independently, bidirectional with probability $θ$, and either of the two possible choices of single direction with probability $\frac12(1-θ)$. The approximation involves two independent copies of a trivariate limiting random vector $(W^*_1,W^*_2,W_3)$ associated with a $3$-type Bienaymé--Galton--Watson process. The approximation error is shown to be typically of order $O(n^{-1/2}\log n)$; this asymptotic order is likely to be optimal, even for the corresponding approximation in the Bernoulli random graph $\mathcal{G}(n,p)$.

math.PR↗

arXiv2026

A Universal Nearest-Neighbor Estimator for Intrinsic Dimensionality

Estimating the intrinsic dimensionality (ID) of data is a fundamental problem in machine learning and computer vision, providing insight into the true degrees of freedom underlying high-dimensional observations. Existing methods often rely on geometric or distributional assumptions and can significantly fail when these assumptions are violated. In this paper, we introduce a novel ID estimator based on nearest-neighbor distance ratios that involves simple calculations and achieves state-of-the-art results. Most importantly, we provide a theoretical analysis proving that our estimator is \emph{universal}, namely, it converges to the true ID independently of the distribution generating the data. We present experimental results on benchmark manifolds and real-world datasets to demonstrate the performance of our estimator.

cs.LG↗

arXiv2026

TASTE: Task-Aware Out-of-Distribution Detection via Stein Operators

Out-of-distribution detection methods are often either data-centric, detecting deviations from the training input distribution irrespective of their effect on a trained model, or model-centric, relying on classifier outputs without explicit reference to data geometry. We propose TASTE (Task-Aware STEin operators): a task-aware framework based on so-called Stein operators, which allows us to link distribution shift to the input sensitivity of the model. We show that the resulting operator admits a clear geometric interpretation as a projection of distribution shift onto the sensitivity field of the model, yielding theoretical guarantees. Beyond detecting the presence of a shift, the same construction enables its localisation through a coordinate-wise decomposition, and for image data-provides interpretable per-pixel diagnostics. Experiments on controlled Gaussian shifts, MNIST under geometric perturbations, and CIFAR-10 perturbed benchmarks demonstrate that the proposed method aligns closely with task degradation while outperforming established baselines.

cs.LG↗

arXiv2025

Split Conformal Prediction under Data Contamination

Conformal prediction is a non-parametric technique for constructing prediction intervals or sets from arbitrary predictive models under the assumption that the data is exchangeable. It is popular as it comes with theoretical guarantees on the marginal coverage of the prediction sets and the split conformal prediction variant has a very low computational cost compared to model training. We study the robustness of split conformal prediction in a data contamination setting, where we assume a small fraction of the calibration scores are drawn from a different distribution than the bulk. We quantify the impact of the corrupted data on the coverage and efficiency of the constructed sets when evaluated on "clean" test points, and verify our results with numerical experiments. Moreover, we propose an adjustment in the classification setting which we call Contamination Robust Conformal Prediction, and verify the efficacy of our approach using both synthetic and real datasets.

stat.ML↗

arXiv2025

FraudTransformer: Time-Aware GPT for Transaction Fraud Detection

Detecting payment fraud in real-world banking streams requires models that can exploit both the order of events and the irregular time gaps between them. We introduce FraudTransformer, a sequence model that augments a vanilla GPT-style architecture with (i) a dedicated time encoder that embeds either absolute timestamps or inter-event values, and (ii) a learned positional encoder that preserves relative order. Experiments on a large industrial dataset -- tens of millions of transactions and auxiliary events -- show that FraudTransformer surpasses four strong classical baselines (Logistic Regression, XGBoost and LightGBM) as well as transformer ablations that omit either the time or positional component. On the held-out test set it delivers the highest AUROC and PRAUC.

cs.LG↗

arXiv2025

Learning to Fuse Temporal Proximity Networks: A Case Study in Chimpanzee Social Interactions

How can we identify groups of primate individuals which could be conjectured to drive social structure? To address this question, one of us has collected a time series of data for social interactions between chimpanzees. Here we use a network representation, leading to the task of combining these data into a time series of a single weighted network per time stamp, where different proximities should be given different weights reflecting their relative importance. We optimize these proximity-type weights in a principled way, using an innovative loss function which rewards structural consistency for consecutive time steps. The approach is empirically validated by carefully designed synthetic data. Using statistical tests, we provide a way of identifying groups of individuals that stay related for a significant length of time. Applying the approach to the chimpanzee data set, we detect cliques in the animal social network time series, which can be validated by real-world intuition from prior research and qualitative observations by chimpanzee experts.

stat.ML↗

arXiv2025

Generalization and Robustness of the Tilted Empirical Risk

The generalization error (risk) of a supervised statistical learning algorithm quantifies its prediction ability on previously unseen data. Inspired by exponential tilting, \citet{li2020tilted} proposed the {\it tilted empirical risk} (TER) as a non-linear risk metric for machine learning applications such as classification and regression problems. In this work, we examine the generalization error of the tilted empirical risk in the robustness regime under \textit{negative tilt}. Our first contribution is to provide uniform and information-theoretic bounds on the {\it tilted generalization error}, defined as the difference between the population risk and the tilted empirical risk, under negative tilt for unbounded loss function under bounded $(1+ε)$-th moment of loss function for some $ε\in(0,1]$ with a convergence rate of $O(n^{-ε/(1+ε)})$ where $n$ is the number of training samples, revealing a novel application for TER under no distribution shift. Secondly, we study the robustness of the tilted empirical risk with respect to noisy outliers at training time and provide theoretical guarantees under distribution shift for the tilted empirical risk. We empirically corroborate our findings in simple experimental setups where we evaluate our bounds to select the value of tilt in a data-driven manner.

stat.ML↗

arXiv2025

Individualised Counterfactual Examples Using Conformal Prediction Intervals

Counterfactual explanations for black-box models aim to pr ovide insight into an algorithmic decision to its recipient. For a binary classification problem an individual counterfactual details which features might be changed for the model to infer the opposite class. High-dimensional feature spaces that are typical of machine learning classification models admit many possible counterfactual examples to a decision, and so it is important to identify additional criteria to select the most useful counterfactuals. In this paper, we explore the idea that the counterfactuals should be maximally informative when considering the knowledge of a specific individual about the underlying classifier. To quantify this information gain we explicitly model the knowledge of the individual, and assess the uncertainty of predictions which the individual makes by the width of a conformal prediction interval. Regions of feature space where the prediction interval is wide correspond to areas where the confidence in decision making is low, and an additional counterfactual example might be more informative to an individual. To explore and evaluate our individualised conformal prediction interval counterfactuals (CPICFs), first we present a synthetic data set on a hypercube which allows us to fully visualise the decision boundary, conformal intervals via three different methods, and resultant CPICFs. Second, in this synthetic data set we explore the impact of a single CPICF on the knowledge of an individual locally around the original query. Finally, in both our synthetic data set and a complex real world dataset with a combination of continuous and discrete variables, we measure the utility of these counterfactuals via data augmentation, testing the performance on a held out set.

stat.ML↗

arXiv2025

Triadic structures in multislice networks

Networks provide a popular representation of complex data. Often, different types of relational measurements are taken on the same subjects. Such data can be represented as a \textit{multislice network}, a collection of networks on the same set of nodes, with connections between the different layers to be determined. For the analysis of multislice networks, we take inspiration from the analysis of simple networks, for which small subgraphs (motifs) have proven to be useful; motifs are even seen as building blocks of complex networks. A particular instance of a motif is a triangle, and while triangle counts are well understood for simple network models such as Erdős-Rényi random graphs, with i.i.d. distributed edges, even for simple multislice network models little is known about triangle counts. Here we address this issue by extending the analysis of triadic structures to multislice Erdős-Rényi networks. Again taking inspiration from the analysis of sparse Erdős-Rényi random graphs, we show that the distribution of triangles across multiple layers in a multislice Erdős-Rényi network can be well approximated by an appropriate Poisson distribution. This theoretical result opens the door to statistical goodness of fit tests for multislice networks.

math.PR↗

arXiv2025

Stein's method for distributions modelling competing and complementary risk problems

Competing and Complementary risk (CCR) problems are often modelled using a class of distributions of the maximum, or minimum, of a random number of i.i.d. random variables; we call this class the CCR class of distributions. While the CCR distributions generally do not have an easy-to-calculate density or probability mass function, two special cases, namely the Poisson-exponential and the exponential geometric distributions, can easily be calculated. Hence, it is of interest to approximate CCR distributions with these simpler distributions. In this paper, we develop Stein's method for the CCR class of distributions to provide a general comparison approach to bound the distance between two CCR distributions and contrast this approach to bounds obtained using a Lindeberg argument. We detail the comparison for Poisson-exponential and exponential-geometric distributions.

math.PR↗

arXiv2025

Private Synthetic Graph Generation and Fused Gromov-Wasserstein Distance

Networks are popular for representing complex data. In particular, differentially private synthetic networks are much in demand for method and algorithm development. The network generator should be easy to implement and should come with theoretical guarantees. Here we start with complex data as input and jointly provide a network representation as well as a synthetic network generator. Using a random connection model, we devise an effective algorithmic approach for generating attributed synthetic graphs which is $ε$-differentially private at the vertex level, while preserving utility under an appropriate notion of distance which we develop. We provide theoretical guarantees for the accuracy of the private synthetic graphs using the fused Gromov-Wasserstein distance, which extends the Wasserstein metric to structured data. Our method draws inspiration from the PSMM method of \citet{he2023}.

stat.ML↗

arXiv2025

Isolated vertices in two duplication-divergence models with edge deletion

Duplication-divergence models are a popular model for the evolution of gene and protein interaction networks. However, existing duplication-divergence models often neglect realistic features such as loss of interactions. Thus, in this paper we present two novel models that incorporate random edge deletions into the duplication-divergence framework. As in protein-protein interaction networks, with proteins as vertices and interactions as edges, by design isolated vertices tend to be rare, our main focus is on the number of isolated vertices; our main result gives lower and upper bounds for the proportion of isolated vertices, when the network size is large. Using these bounds we identify the parameter regimes for which almost all vertices are typically isolated; and also show that there are parameter regimes in which the proportion of isolated vertices can be bounded away from 0 and 1 with high probability. In addition, we find regimes in which the proportion of isolated vertices tends to be small. The proof relies on a standard martingale argument, which in turn requires a careful analysis of the first two moments of the expected degree distribution. The theoretical findings are illustrated by simulations, indicating that as the network size tends to infinity, the proportion of isolated vertices can converge to a limit that is neither 0 or 1.

math.PR↗

arXiv2025

Normal approximation for the posterior in exponential families

In this paper, we obtain quantitative, non-asymptotic, and data-dependent \textit{Bernstein-von Mises type} bounds on the normal approximation of the posterior distribution in exponential family models with arbitrary centring and scaling. Our bounds, stated in the total variation and Wasserstein distances, are valid for univariate and multivariate posteriors alike, and do not require a conjugate prior setting. They are obtained through a refined version of Stein's method of comparison of operators that allows for improved dimensional dependence in high-dimensional settings and may also be of interest in other problems. Our approach is rather flexible and, in certain settings, allows for the derivation of bounds with rates of convergence faster than the usual $ O(n^{-1/2}) $ rate (when $ n $ is the sample size). We illustrate our findings on a variety of exponential family distributions, including the Weibull, multinomial, and linear regression with unknown variance. The resulting bounds have an explicit dependence on the prior distribution and on sufficient statistics of the data from the sample, and thus provide insight into how these factors affect the quality of the normal approximation. Insights from our examples include identification of conditions under which faster $ O(n^{-1}) $ convergence rates occur for Bernoulli data, illustrations of how the quality of the normal approximation is influenced by the choice of standardisation, and dimensional dependence in high-dimensional settings.

math.ST↗