Searcharxiv⌕ Search

arXiv subjects

Gaia Carenini

Publications and source records attributed to Gaia Carenini.

At least 19 recordsLinked to original sources
arXiv2026

Ranked spreadness and sample-based testing

In this note, we introduce the notion of ranked spreadness, a strengthening of the usual spread condition in which the elements of each member can be ordered so that their one-coordinate marginals decay geometrically with their rank. This additional structure removes the dependence on the maximum set size in random-containment estimates. We prove width-free hitting and weighted-concentration theorems for ranked-spread set systems, together with an elementary kernel-extraction theorem showing that ranked spreadness arises naturally in arbitrary distributions on small sets. Our main application is to the simulation of nonadaptive property testers by sample-based testers. If a one-sided tester has average query complexity $d$ and rejects every far input with probability at least $δ$, then, for every integer $c>d/δ$, it admits a one-sided sample-based simulation with expected sample complexity $O_{d,δ,|Σ|}\bigl(n^{1-1/c}\bigr)$. More generally, if positive inputs are rejected with probability at most $γ$ and far inputs with probability at least $δ>γ$, the same conclusion holds for every $c>d/(δ-γ)$. In particular, for constant-query nonadaptive testers we obtain an exponent $1-Θ(1/q)$, matching, up to the dependence on the rejection gap, the exponent conjectured by Fischer, Lachish, and Vasudev.

math.ST↗
arXiv2026

A quantitative container characterization of one-sided testability

We give a quantitative combinatorial characterization of size-oblivious one-sided testability in the dense graph model, resolving a question of Alon, Fischer, Newman, and Shapira. For hereditary graph properties, we prove that one-sided testability is quantitatively equivalent to the existence of suitable hypergraph containers, a central and widely used tool in modern combinatorics. Combining this equivalence with the Alon-Shapira notion of semi-hereditariness yields a quantitative characterization of arbitrary graph properties. The correspondence is effective in both directions and provides explicit translations between tester complexity and container parameters. Our proof is regularity-free and extends uniformly to every fixed finite relational signature of bounded arity, including digraphs, coloured graphs, and hypergraphs. As applications, we obtain quantitative closure results for partition properties and testers for properties defined by the existence of a linearly large induced substructure.

math.CO↗
arXiv2026

Embedding Induced Bounded Degree Graphs

We prove a sparse embedding theorem for induced embeddings of bounded-degree graphs. The theorem applies to pairs $G\subseteq Γ$: the graph $G$ supplies the positive edges of the target graph, while the ambient graph $Γ$ supplies the induced constraints which must be avoided. Its main feature is that these two types of constraints are kept separate throughout the embedding process. As an application, we show that, for every fixed $Δ,r\ge2$, there are constants $A,C>0$ such that every $n$-vertex graph $H$ with maximum degree at most $Δ$ satisfies $r_{\text{ind}}(H;r)\le C n^{Δ+2}(\log n)^A$. This improves the exponent in the polynomial bound of Conlon, Fox and Zhao for bounded-degree induced Ramsey numbers. The proof combines the aforementioned embedding theorem with a sparse random transference argument, in which the random host is used only to certify robust deterministic hypotheses for every colour class.

math.CO↗
arXiv2026

A unified abstract regularity lemma

The goal of this short note is to prove a unified abstract regularity lemma which recovers Szemerédi's graph regularity lemma, Green's arithmetic regularity lemma, and a regularity lemma for Boolean functions as direct corollaries.

math.CO↗
arXiv2026

A Sparse Transference Principle for a Non-Monotone Ramsey Property

We prove a sparse transference theorem for induced Ramsey graphs. The theorem transfers the weighted random-host proof of Aragão, Campos, Dahia, Filipe, and Marciano to the sparse random setting. It follows that, for every fixed graph $H$ with no isolated vertices and at least two edges, and every $η>0$, there is $C>0$ such that, whenever $N\ge r^{Cr}$ and $N^{-1/m_2(H)+η}\le p\le \frac12$, with high probability every $r$-colouring of the edges of $G(N,p)$ contains a monochromatic induced copy of $H$. Here, $m_2(H)$ denotes the usual maximum 2-density of $H$.

math.CO↗
arXiv2026

A strengthening of Chang's lemma

We prove a strengthening of Chang's lemma for subsets of $\mathbb F_p^n$. The classical conclusion that the large spectrum is contained in a subspace of dimension at most $2\varepsilon^{-2}\log(1/α)$ is refined to show that every character outside this subspace has small correlation with the set not only globally, but also on average over the cosets of the orthogonal complement, in a natural cosetwise $\ell^1$ norm. As a consequence, we obtain a localized counting lemma. We also give an extension of the argument to arbitrary finite abelian groups.

math.NT↗
arXiv2025

On Modular Edge Colourings of Graphs

Given a graph $G$ and an integer $k\geq 2$, let $χ'_k(G)$ denote the minimum number of colours required to colour the edges of $G$ such that, in each colour class, the subgraph induced by the edges of that colour has all non-zero degrees congruent to $1$ modulo $k$. In 1992, Pyber proved that $χ'_2(G) \leq 4$ for every graph $G$, and posed the question of whether $χ'_k(G)$ can be bounded solely in terms of $k$ for every $k\geq 3$. This question was answered in 1997 by Scott, who showed that $χ'_k(G)\leq5k^2\log k$, and further asked whether $χ'_k(G) = O(k)$. Recently, Botler, Colucci, and Kohayakawa (2023) answered Scott's question affirmatively proving that $χ'_k(G) \leq 198k - 101$, and conjectured that the multiplicative constant could be reduced to $1$. A step towards this latter conjecture was made in 2024 by Nweit and Yang, who improved the bound to $χ'_k(G) \leq 177k - 93$. In this paper, we further improve the multiplicative constant to $9$. More specifically, we prove that there is a function $f\in o(k)$ for which $χ'_k(G) \leq 7k + f(k)$ if $k$ is odd, and $χ'_k(G) \leq 9k + f(k)$ if $k$ is even. In doing so, we prove that $χ'_k(G) \leq k + O(d)$ for every $d$-degenerate graph $G$, which plays a central role in our proof.

math.CO↗
arXiv2025

Quantum Automating $\mathbf{TC}^0$-Frege Is LWE-Hard

We prove the first hardness results against efficient proof search by quantum algorithms. We show that under Learning with Errors (LWE), the standard lattice-based cryptographic assumption, no quantum algorithm can weakly automate $\mathbf{TC}^0$-Frege. This extends the line of results of Krajíček and Pudlák (Information and Computation, 1998), Bonet, Pitassi, and Raz (FOCS, 1997), and Bonet, Domingo, Gavaldà, Maciel, and Pitassi (Computational Complexity, 2004), who showed that Extended Frege, $\mathbf{TC}^0$-Frege and $\mathbf{AC}^0$-Frege, respectively, cannot be weakly automated by classical algorithms if either the RSA cryptosystem or the Diffie-Hellman key exchange protocol are secure. To the best of our knowledge, this is the first interaction between quantum computation and propositional proof search.

cs.CC↗
arXiv2025

On the Modular Chromatic Index of Random Hypergraphs

Let $k,r \geq 2$ be two integers. We consider the problem of partitioning the hyperedge set of an $r$-uniform hypergraph $H$ into the minimum number $χ_k'(H)$ of edge-disjoint subhypergraphs in which every vertex has either degree $0$ or degree congruent to $1$ modulo $k$. For a random hypergraph $H$ drawn from the binomial model $\mathbf{H}(n,p,r)$, with edge probability $p \in (C\log(n)/n,1)$ for a large enough constant $C>0$ independent of $n$ and satisfying $n^{r-1}p(1-p)\to\infty$ as $n\to\infty$, we show that asymptotically almost surely $χ_k'(H) = k$ if $n$ is divisible by $\gcd(k,r)$, and $\max(k,r) \le χ_k'(H) \le k+r+1$ otherwise. A key ingredient in our approach is a sufficient condition ensuring the existence of a $k$-factor, a $k$-regular spanning subhypergraph, within subhypergraphs of a random hypergraph from $\mathbf{H}(n,p,r)$, a result that may be of independent interest. Our main result extends a theorem of Botler, Colucci, and Kohayakawa (2023), who proved an analogous statement for graphs, and provides a partial answer to a question posed by Goetze, Klute, Knauer, Parada, Peña, and Ueckerdt (2025) regarding whether $χ_2'(H)$ can be bounded by a constant for every hypergraph $H$.

math.CO↗
arXiv2025

Stellar flare detection in XMM-Newton with gradient boosted trees

The EXTraS project, based on data collected with the XMM-Newton observatory, provided us with a vast amount of light curves for X-ray sources. For each light curve, EXTraS also provided us with a set of features (https://extras.inaf.it). We extract from the EXTraS database a tabular dataset of 31,832 variable sources by 108 features. Of these, 13,851 sources were manually labeled as stellar flares or non-flares based on direct visual inspection. We employ a supervised learning approach to produce a catalog of stellar flares based on our dataset, releasing it to the community. We leverage explainable AI tools and interpretable features to better understand our classifier. We train a gradient boosting classifier on 80\% of the data for which labels are available. We compute permutation feature importance scores, visualize feature space using UMAP, and analyze some false positive and false negative data points with the help of Shapley additive explanations -- an AI explainability technique used to measure the importance of each feature in determining the classifier's prediction for each instance. On the test set made up of the remainder 20\% of our labeled data, we obtain an accuracy of 97.1\%, with a precision of 82.4\% and a recall of 73.3\%. Our classifier outperforms a simple criterion based on fitting the light curve with a flare template and significantly surpasses a gradient-boosted classifier trained only on model-independent features. False positives appear related to flaring light curves that are not associated with a stellar counterpart, while false negatives often correspond to multiple flares or otherwise peculiar or noisy curves. We apply our trained classifier to currently unlabeled sources, releasing the largest catalog of X-ray stellar flares to date. [abridged]

astro-ph.HE↗
arXiv2025

Property Testing in Bounded Degree Hypergraphs

We extend the bounded degree graph model for property testing introduced by Goldreich and Ron (Algorithmica, 2002) to hypergraphs. In this framework, we analyse the query complexity of three fundamental hypergraph properties: colorability, $k$-partiteness, and independence number. We present a randomized algorithm for testing $k$-partiteness within families of $k$-uniform $n$-vertex hypergraphs of bounded treewidth whose query complexity does not depend on $n$. In addition, we prove optimal lower bounds of $Ω(n)$ on the query complexity of testing algorithms for $k$-colorability, $k$-partiteness, and independence number in $k$-uniform $n$-vertex hypergraphs of bounded degree. For each of these properties, we consider the problem of explicitly constructing $k$-uniform hypergraphs of bounded degree that differ in $Θ(n)$ hyperedges from any hypergraph satisfying the property, but where violations of the latter cannot be detected in any neighborhood of $o(n)$ vertices.

cs.CC↗
arXiv2024

Variable Star Light Curves in Koopman Space

We present the first application of data-driven techniques for dynamical system analysis based on Koopman theory to variable stars. We focus on light curves of RRLyrae type variables, in the Galactic globular cluster $ω$ Centauri. Light curves are thus summarized by a handful of complex eigenvalues, corresponding to oscillatory or fading dynamical modes. We find that variable stars of the RRc subclass can be summarized in terms of fewer ($\approx 8$) eigenvalues, while RRab need comparatively more ($\approx 12$). This result can be leveraged for classification and reflects the simpler structure of RRc light curves. We then consider variable stars displaying secular variations due to the Tseraskaya-Blazhko effect and find a change in relevant eigenvalues with time, with possible implications for the physical interpretation of the effect.

astro-ph.SR↗
arXiv2024

Towards a Fully Interpretable and More Scalable RSA Model for Metaphor Understanding

The Rational Speech Act (RSA) model provides a flexible framework to model pragmatic reasoning in computational terms. However, state-of-the-art RSA models are still fairly distant from modern machine learning techniques and present a number of limitations related to their interpretability and scalability. Here, we introduce a new RSA framework for metaphor understanding that addresses these limitations by providing an explicit formula - based on the mutually shared information between the speaker and the listener - for the estimation of the communicative goal and by learning the rationality parameter using gradient-based methods. The model was tested against 24 metaphors, not limited to the conventional $\textit{John-is-a-shark}$ type. Results suggest an overall strong positive correlation between the distributions generated by the model and the interpretations obtained from the human behavioral data, which increased when the intended meaning capitalized on properties that were inherent to the vehicle concept. Overall, findings suggest that metaphor processing is well captured by a typicality-based Bayesian model, even when more scalable and interpretable, opening up possible applications to other pragmatic phenomena and novel uses for increasing Large Language Models interpretability. Yet, results highlight that the more creative nuances of metaphorical meaning, not strictly encoded in the lexical concepts, are a challenging aspect for machines.

cs.CL↗
arXiv2023

Interpretable machine learning for finding intermediate-mass black holes

Definitive evidence that globular clusters (GCs) host intermediate-mass black holes (IMBHs) is elusive. Machine learning (ML) models trained on GC simulations can in principle predict IMBH host candidates based on observable features. This approach has two limitations: first, an accurate ML model is expected to be a black box due to complexity; second, despite our efforts to realistically simulate GCs, the simulation physics or initial conditions may fail to fully reflect reality. Therefore our training data may be biased, leading to a failure in generalization on observational data. Both the first issue -- explainability/interpretability -- and the second -- out of distribution generalization and fairness -- are active areas of research in ML. Here we employ techniques from these fields to address them: we use the anchors method to explain an XGBoost classifier; we also independently train a natively interpretable model using Certifiably Optimal RulE ListS (CORELS). The resulting model has a clear physical meaning, but loses some performance with respect to XGBoost. We evaluate potential candidates in real data based not only on classifier predictions but also on their similarity to the training data, measured by the likelihood of a kernel density estimation model. This measures the realism of our simulated data and mitigates the risk that our models may produce biased predictions by working in extrapolation. We apply our classifiers to real GCs, obtaining a predicted classification, a measure of the confidence of the prediction, an out-of-distribution flag, a local rule explaining the prediction of XGBoost and a global rule from CORELS.

astro-ph.GA↗
arXiv2023

Tackling Computational Heterogeneity in FL: A Few Theoretical Insights

The future of machine learning lies in moving data collection along with training to the edge. Federated Learning, for short FL, has been recently proposed to achieve this goal. The principle of this approach is to aggregate models learned over a large number of distributed clients, i.e., resource-constrained mobile devices that collect data from their environment, to obtain a new more general model. The latter is subsequently redistributed to clients for further training. A key feature that distinguishes federated learning from data-center-based distributed training is the inherent heterogeneity. In this work, we introduce and analyse a novel aggregation framework that allows for formalizing and tackling computational heterogeneity in federated optimization, in terms of both heterogeneous data and local updates. Proposed aggregation algorithms are extensively analyzed from a theoretical, and an experimental prospective.

cs.LG↗
arXiv2023

Sparse logistic regression for RR Lyrae vs binaries classification

RR Lyrae (RRL) are old, low-mass radially pulsating variable stars in their core helium burning phase. They are popular stellar tracers and primary distance indicators, since they obey to well defined period-luminosity relations in the near-infrared regime. Their photometric identification is not trivial, indeed, RRL samples can be contaminated by eclipsing binaries, especially in large datasets produced by fully automatic pipelines. Interpretable machine-learning approaches for separating eclipsing binaries from RRL are thus needed. Ideally, they should be able to achieve high precision in identifying RRL while generalizing to new data from different instruments. In this paper, we train a simple logistic regression classifier on Catalina Sky Survey (CSS) light curves. It achieves a precision of 87% at 78% recall for the RRL class on unseen CSS light curves. It generalizes on out-of-sample data (ASAS/ASAS-SN light curves) with a precision of 85% at 96% recall. We also considered a L1-regularized version of our classifier, which reaches 90% sparsity in the light-curve features with a limited trade-off in accuracy on our CSS validation set and -- remarkably -- also on the ASAS/ASAS-SN light curve test set. Logistic regression is natively interpretable, and regularization allows us to point out the parts of the light curves that matter the most in classification. We thus achieved both good generalization and full interpretability.

astro-ph.SR↗
arXiv2023

Adversarial Path Planning for Optimal Camera Positioning

The use of visual sensors is flourishing, driven among others by the several applications in detection and prevention of crimes or dangerous events. While the problem of optimal camera placement for total coverage has been solved for a decade or so, that of the arrangement of cameras maximizing the recognition of objects "in-transit" is still open. The objective of this paper is to attack this problem by providing an adversarial method of proven optimality based on the resolution of Hamilton-Jacobi equations. The problem is attacked by first assuming the perspective of an adversary, i.e. computing explicitly the path minimizing the probability of detection and the quality of reconstruction. Building on this result, we introduce an optimality measure for camera configurations and perform a simulated annealing algorithm to find the optimal camera placement.

cs.CG↗
arXiv2022

Federated Learning Aggregation: New Robust Algorithms with Guarantees

Federated Learning has been recently proposed for distributed model training at the edge. The principle of this approach is to aggregate models learned on distributed clients to obtain a new more general "average" model (FedAvg). The resulting model is then redistributed to clients for further training. To date, the most popular federated learning algorithm uses coordinate-wise averaging of the model parameters for aggregation. In this paper, we carry out a complete general mathematical convergence analysis to evaluate aggregation strategies in a federated learning framework. From this, we derive novel aggregation algorithms which are able to modify their model architecture by differentiating client contributions according to the value of their losses. Moreover, we go beyond the assumptions introduced in theory, by evaluating the performance of these strategies and by comparing them with the one of FedAvg in classification tasks in both the IID and the Non-IID framework without additional hypothesis.

stat.ML↗