Searcharxiv⌕ Search

arXiv subjects

Francesco Regazzoni

Publications and source records attributed to Francesco Regazzoni.

At least 19 recordsLinked to original sources

Fault Injection Attacks on Q-Learning: Analysis, Vulnerabilities, and Countermeasures

Hardware fault injection attacks threaten machine learning systems, yet their impact on reinforcement learning (RL) has received little attention despite the use of RL in safety-critical applications. This paper presents a security analysis of Q-Learning implementations under voltage and clock glitching attacks on a ChipWhisperer platform, evaluating five variants: Standard, Robust, Resilient, Linear, and Deep Q-Learning. In standard Q-Table implementations, faults corrupt reward computation, action selection, and Q-value updates during learning, and change the selected action during inference. Robust Q-Learning does not prevent these attacks: NaN and infinite values injected during learning persist in the Q-table, and its greedy policy uses the same code as standard Q-Learning. This paper also evaluates a Resilient Q-Learning variant that applies triple modular redundancy with majority voting. It corrects corruption of a single Q-table copy, but inputs shared by the copies, such as the reward, and the voting code itself remain outside its protection. The analysis extends to Linear and Deep Q-Learning, where faults corrupt stored weights during learning and change actions during inference. Moving beyond benchmark environments, this work evaluates an online traffic light controller in a microscopic traffic simulator, where a single injected infinite value degrades accumulated vehicle waiting time by 148x while corrupting under 4% of the Q-table, and where triple modular redundancy corrected the faults injected in our tests, although glitching could still halt the device. Altogether, this work provides a basis for understanding and securing RL deployments against fault injection attacks.

cs.CR↗

Learning PDE solution operators with variable initial conditions via Latent Dynamics Networks

In many-query scenarios, data-driven surrogate models provide an efficient alternative to high-fidelity solvers for simulating physical systems governed by Partial Differential Equations (PDEs). In this context, the Latent Dynamics Network (LDNet) has recently demonstrated remarkable performance in predicting the response of spatio-temporal systems, combining Neural Ordinary Differential Equations with nonlinear dimensionality reduction. However, the original formulation assumes a fixed initial condition, limiting its applicability to many real-world applications where a system evolves from varying starting states. In this work, we overcome this limitation while keeping the end-to-end training procedure of the original LDNet and its encoder-free nature, which preserves its intrinsic independence from spatial resolution and grid topology. We infer the initial latent state directly from a small set of early-time observations, treating latent-state initialization as an adaptation problem, and investigate two strategies: an auto-decoding formulation and a meta-learning approach in which the initial latent state acts as a task-specific context variable. We demonstrate the accuracy of the proposed methods across diverse physical phenomena, spanning advection-diffusion, fluid dynamics, and solid mechanics. Meta-learning markedly accelerates latent-state inference and induces smoother, better-conditioned optimization landscapes, and spontaneously organizes the latent space into a structured representation that reflects physically meaningful features of the underlying dynamics. The coordinate-based decoder enables training from spatially subsampled data while recovering high-resolution solution fields at inference. The resulting approach provides an efficient and resolution-independent surrogate modeling framework for many-query simulations of time-dependent PDEs with varying initial conditions.

cs.LG↗

Optimal Transport Dropout for Structured Predictive Uncertainty

Deterministic neural networks and neural operators provide point predictions with no intrinsic measure of reliability. Yet, predictive uncertainty may stem from irreducible outcome variability, finite data, or limitations of the chosen model class. Monte Carlo dropout offers a computationally convenient way to construct a predictive distribution through stochastic feature masking, without training multiple independent networks or explicitly inferring a posterior over model parameters. However, its perturbation law is largely prescribed a priori and typically factorised across latent coordinates. We introduce Optimal Transport Dropout (OTD), which instead learns the predictive mapping and the law of its latent perturbations jointly. Starting from a simple independent reference distribution, OTD transports latent perturbations through a learnable flow and propagates them through the predictive neural network, thereby inducing a structured predictive law. Training uses the strictly proper Energy Score, while a kinetic-action term geometrically regularises the transport. Synthetic benchmarks show that OTD captures multimodal predictive distributions, generates meaningful dispersion when the model is misspecified, and exhibits contracting dispersion as more training data or greater model capacity are provided. For a field-valued partial differential equation surrogate, predictive dispersion strongly aligns with the spatial pattern of prediction errors. On this task, compared with Monte Carlo dropout, OTD yields more accurate predictions and better-calibrated, substantially narrower intervals. On real-world regression benchmarks, it further shows competitive accuracy and better probabilistic predictions compared to established baselines. OTD therefore offers a way to learn structured predictive uncertainty without explicit posterior inference or ensembles of independently trained predictors.

cs.LG↗

A mathematical model for irreversible damage of the collagen scaffold in the myocardium

We propose a dissipative, irreversible damage model for anisotropic media in large deformations to address the damage process of the myocardium following a cardiac infarction. We model damage to the collagen in the cleavage planes resulting from an increased load to the passive tissue. Starting from variational principles, we derive a quasi-static differential model governing the irreversible evolution of the damage. We apply our model to two test cases. First, a rectangular passive slab geometry to which an indenter-like load is applied. Then, we couple our model with a model for the numerical simulation of left ventricular electromechanics. Our results show that a redistribution of the load within a left ventricle occurring following a cardiac infarction can produce the damage to the collagen scaffold in the infarcted area consistent with experimental results. This is indicative of the fact that passive load-dependent dissipative phenomena are implicated in post-infarction remodelling.

math.NA↗

Hyperelastic constitutive model discovery with differentiable finite elements and structure-preserving neural networks

The discovery of constitutive laws from experimentally accessible measurements is a central problem in nonlinear computational mechanics. Many data-driven constitutive identification approaches rely either on paired strain-stress data or on full-field displacement measurements, both of which are difficult to obtain in realistic three-dimensional settings. We present a differentiable finite element framework for the discovery of hyperelastic material laws from partial observations, including boundary-only displacement measurements and global reaction forces. The method embeds the nonlinear finite element equilibrium problem directly into the learning loop, so that candidate strain-energy densities are assessed through the deformation fields and reactions they induce. This formulation enforces mechanical equilibrium as a constraint and allows the loss function to be evaluated only at observed locations. To ensure physical admissibility and promote numerical solvability throughout training, the constitutive response is represented by Hyperelastic Neural Networks, a structure-preserving neural class that enforces residual energy and stress-free conditions, frame indifference, isotropic material symmetry, polyconvexity, coercivity, and controlled volumetric growth by construction. The resulting PDE-constrained learning problem is solved using a quasi-Newton strategy combined with continuation and solver-aware backtracking. Numerical experiments in two- and three-dimensional finite elasticity demonstrate accurate recovery of hyperelastic isotropic responses from boundary-only data, robustness to measurement noise, and generalization across geometries, loading conditions, and boundary conditions.

math.NA↗

A stability-preserving polytopal discontinuous Galerkin method for the Fisher-Kolmogorov model with applications to neurodegenerative disease modelling

The Fisher--Kolmogorov equation models the spatio-temporal evolution of interacting biological species and is extensively employed in fields such as ecology, population dynamics, and the modelling of neurodegenerative diseases. Under suitable assumptions on the data, the solution $c$ is non-negative, a key feature because $c$ typically denotes a population density or a relative concentration. However, standard discretisation methods often fail to preserve this property, resulting in non-physical oscillations and unstable numerical approximations. In this work, we propose and analyse a numerical method to stabilise the dynamics of the Fisher--Kolmogorov model around the unstable equilibrium $c=0$. The proposed approach combines a discontinuous Galerkin spatial discretisation on general polygonal and polyhedral meshes with the Crank--Nicolson time integration scheme. The main idea is to suitably modify the formulation at the continuous level so that, on the one hand, it is strongly consistent with the original model, and, on the other hand, it ensures stability when moving to the discrete setting. We prove well-posedness of the semi-discrete formulation, derive stability bounds and prove optimal \textit{a priori} error estimates in a suitable energy norm. The theoretical results are demonstrated through a comprehensive set of numerical examples. Moreover, we consider an application arising in computational neuroscience by simulating the propagation of $α$-synuclein, a key pathogenic protein implicated in Parkinson's disease and other neurodegenerative diseases, demonstrating that the proposed scheme is stable, high-order accurate, and robust in a biologically relevant computational setting.

math.NA↗

Emerging Threats and Countermeasures in Neuromorphic Systems: A Survey

Neuromorphic computing mimics brain-inspired mechanisms through spiking neurons and energy-efficient processing, offering a pathway to efficient in-memory computing (IMC). However, these advancements raise critical security and privacy concerns. As the adoption of bio-inspired architectures and memristive devices increases, so does the urgency to assess the vulnerability of these emerging technologies to hardware and software attacks. Emerging architectures introduce new attack surfaces, particularly due to asynchronous, event-driven processing and stochastic device behavior. The integration of memristors into neuromorphic hardware and software implementations in spiking neural networks offers diverse possibilities for advanced computing architectures, including their role in security-aware applications. This survey systematically analyzes the security landscape of neuromorphic systems, covering attack methodologies, side-channel vulnerabilities, and countermeasures. We focus on both hardware and software concerns relevant to spiking neural networks (SNNs) and hardware primitives, such as Physical Unclonable Functions (PUFs) and True Random Number Generators (TRNGs) for cryptographic and secure computation applications. We approach this analysis from diverse perspectives, from attack methodologies to countermeasure strategies that integrate efficiency and protection in brain-inspired hardware. This review not only maps the current landscape of security threats but provides a foundation for developing secure and trustworthy neuromorphic architectures.

cs.CR↗

Physics-constrained identification of graph-based thermal networks for spacecraft digital twins

Reconstructing a thermal model capable of efficiently simulating the behavior of a spacecraft from sparse and localized temperature measurements remains a challenging task. To address this, we introduce a physically-constrained calibration framework for Lumped Parameter Thermal Models (LPTMs), formulated as a trajectory-based inverse problem for graph dynamical systems. The model reconstructs thermal dynamics directly from temperature measurements and known inputs, without relying on a priori parameter values derived from material properties or geometric assumptions. Physical admissibility is enforced at the parameterization level: positivity of nodal coefficients and symmetry of conductive interactions are imposed by construction. This guarantees stable dynamics and restricts the identification problem to a physically meaningful parameter space, improving conditioning without the need of additional regularization. The identification problem is addressed through trajectory matching, ensuring stable rollout over extended time horizons. The methodology is validated on synthetic datasets generated from high-fidelity finite element simulations under progressively complex forcing conditions. The calibrated LPTMs accurately reproduce long-term temperature evolution and exhibit robustness to measurement noise. The proposed framework provides a systematic approach to the calibration of reduced-order thermal models by combining physical structure with data-driven identification. The numerical results show a favorable balance between accuracy and computational efficiency, making the models suitable for integration in spacecraft thermal Digital Twin applications.

math.NA↗

The functional impact of myofiber macroscopic organization and disarray in computational models of the murine heart

A major challenge in computational models of cardiac electromechanics is the reconstruction of myocardial fiber architecture, as direct in vivo measurements of fiber orientation are not feasible. Consequently, rule-based methods are commonly adopted as surrogates. This study investigates the respective roles of macroscopic fiber architecture and microscopic fiber disarray in cardiac electromechanical simulations. A high-fidelity biventricular electromechanical model of a murine heart was developed using a high-resolution myocardial fiber field obtained via mesoscopic optical imaging, which serves as a reference ground truth. A spatial smoothing strategy is introduced to decouple macroscopic fiber organization from local disarray, and the resulting responses are also compared with those obtained using a rule-based fiber field. The results show that passive mechanics and electrophysiological activation are only weakly affected by fiber disarray, with global chamber compliance and activation times remaining largely unchanged across different fiber descriptions. In contrast, active mechanics is highly sensitive to fiber architecture. Moderate regularization of the experimentally measured fiber field enhances the ventricular pumping efficiency of the computational model by reducing microscopic disarray while preserving the macroscopic helical organization, whereas excessive smoothing or rule-based fiber reconstructions lead to unphysiologically strong or inefficient contraction. Within this framework, two commonly adopted surrogate strategies to account for fiber disarray are investigated: a reduction of the effective cross-bridge stiffness in the active tension model, and the introduction of controlled misalignment between active tension and the local fiber direction. Overall, the results reveal important limitations of commonly adopted surrogate approaches for modeling fiber disarray.

math.NA↗

Learning geometry-dependent lead-field operators for forward ECG modeling

Modern forward electrocardiogram (ECG) computational models rely on an accurate representation of the torso domain. The lead-field method enables fast ECG simulations while preserving full geometric fidelity. Achieving high anatomical accuracy in torso representation is, however, challenging in clinical practice, as imaging protocols are typically focused on the heart and often do not include the entire torso. In addition, the computational cost of the lead-field method scales linearly with the number of electrodes, limiting its applicability in high-density recording settings. To date, no existing approach simultaneously achieves high anatomical fidelity, low data requirements and computational efficiency. In this work, we propose a shape-informed surrogate model of the lead-field operator that serves as a drop-in replacement for the full-order model in forward ECG simulations. The proposed framework consists of two components: a geometry-encoding module that maps anatomical shapes into a low-dimensional latent space, and a geometry-conditioned neural surrogate that predicts lead-field gradients from spatial coordinates, electrode positions and latent codes. The proposed method achieves high accuracy in approximating lead fields both within the torso (mean angular error 5°) and inside the heart, resulting in highly accurate ECG simulations (relative mean squared error <2.5%. The surrogate consistently outperforms the widely used pseudo lead-field approximation while preserving negligible inference cost. Owing to its compact latent representation, the method does not require a fully detailed torso segmentation and can therefore be deployed in data-limited settings while preserving high-fidelity ECG simulations.

cs.LG↗

Elimination-compensation pruning for fully-connected neural networks

The unmatched ability of Deep Neural Networks in capturing complex patterns in large and noisy datasets is often associated with their large hypothesis space, and consequently to the vast amount of parameters that characterize model architectures. Pruning techniques affirmed themselves as valid tools to extract sparse representations of neural networks parameters, carefully balancing between compression and preservation of information. However, a fundamental assumption behind pruning is that expendable weights should have small impact on the error of the network, while highly important weights should tend to have a larger influence on the inference. We argue that this idea could be generalized; what if a weight is not simply removed but also compensated with a perturbation of the adjacent bias, which does not contribute to the network sparsity? Our work introduces a novel pruning method in which the importance measure of each weight is computed considering the output behavior after an optimal perturbation of its adjacent bias, efficiently computable by automatic differentiation. These perturbations can be then applied directly after the removal of each weight, independently of each other. After deriving analytical expressions for the aforementioned quantities, numerical experiments are conducted to benchmark this technique against some of the most popular pruning strategies, demonstrating an intrinsic efficiency of the proposed approach in very diverse machine learning scenarios. Finally, our findings are discussed and the theoretical implications of our results are presented.

cs.LG↗

Shape-informed cardiac mechanics surrogates in data-scarce regimes via geometric encoding and generative augmentation

High-fidelity computational models of cardiac mechanics provide mechanistic insight into the heart function but are computationally prohibitive for routine clinical use. Surrogate models can accelerate simulations, but generalization across diverse anatomies is challenging, particularly in data-scarce settings. We propose a two-step framework that decouples geometric representation from learning the physics response, to enable shape-informed surrogate modeling under data-scarce conditions. First, a shape model learns a compact latent representation of left ventricular geometries. The learned latent space effectively encodes anatomies and enables synthetic geometries generation for data augmentation. Second, a neural field-based surrogate model, conditioned on this geometric encoding, is trained to predict ventricular displacement under external loading. The proposed architecture performs positional encoding by using universal ventricular coordinates, which improves generalization across diverse anatomies. Geometric variability is encoded using two alternative strategies, which are systematically compared: a PCA-based approach suitable for working with point cloud representations of geometries, and a DeepSDF-based implicit neural representation learned directly from point clouds. Overall, our results, obtained on idealized and patient-specific datasets, show that the proposed approaches allow for accurate predictions and generalization to unseen geometries, and robustness to noisy or sparsely sampled inputs.

cs.LG↗

120 Domain-Specific Languages for Security

Security engineering, from security requirements engineering to the implementation of cryptographic protocols, is often supported by domain-specific languages (DSLs). Unfortunately, a lack of knowledge about these DSLs, such as which security aspects are addressed and when, hinders their effective use and further research. This systematic literature review examines 120 security-oriented DSLs based on six research questions concerning security aspects and goals, language-specific characteristics, integration into the software development lifecycle (SDLC), and effectiveness of the DSLs. We observe a high degree of fragmentation, which leads to opportunities for integration. We also need to improve the usability and evaluation of security DSLs.

cs.CR↗

Cardiocirculatory Computational Models for the Study of Hypertension

In this work, we develop patient-specific cardiocirculatory models with the aim of building Digital Twins for hypertension. In particular, in our pathophysiology-based framework, we consider both 0D cardiocirculatory models and a 3D-0D electromechanical model. The 0D model, which consists of an RLC circuit, is studied in two variants, with and without capillaries. The 3D-0D model consists of a three-dimensional electromechanical model of the left ventricle, coupled with a 0D model for the external blood circulation: this representation enables the assessment of additional quantities related to ventricular deformation and stress, and offers a more detailed representation compared to a fully 0D model. Sensitivity analysis is performed on the 0D model, with both a mono- and a multi-parametric approach, in order to identify the parameters that most influence the model outputs and guide the calibration process. We studied three different scenarios, corresponding to systemic, pulmonary and renovascular hypertension, each in three nuances of severity. To maintain a fair comparison among the models, a parameter calibration strategy is developed; the outputs of the 0D model with capillaries are utilized to enhance the 3D-0D model. The results demonstrate that the 3D-0D model yields an accurate representation of cardiocirculatory dynamics in the presence of hypertension; this model represents a powerful step toward digital twins for real-time hypertension control, providing refined and clinically meaningful insights beyond those achievable with 0D models alone.

math.NA↗

Improvements on uncertainty quantification with variational autoencoders

Inverse problems aim to determine model parameters of a mathematical problem from given observational data. Neural networks can provide an efficient tool to solve these problems. In the context of Bayesian inverse problems, Uncertainty Quantification Variational AutoEncoders (UQ-VAE), a class of neural networks, approximate the posterior distribution mean and covariance of model parameters. This allows for both the estimation of the parameters and their uncertainty in relation to the observational data. In this work, we propose a novel loss function for training UQ-VAEs, which includes, among other modifications, the removal of a sample mean term from an already existing one. This modification improves the accuracy of UQ-VAEs, as the original theoretical result relies on the convergence of the sample mean to the expected value (a condition that, in high dimensional parameter spaces, requires a prohibitively large number of samples due to the curse of dimensionality). Avoiding the computation of the sample mean significantly reduces the training time in high dimensional parameter spaces compared to previous literature results. Under this new formulation, we establish a new theoretical result for the approximation of the posterior mean and covariance for general mathematical problems. We validate the effectiveness of UQ-VAEs through three benchmark numerical tests: a Poisson inverse problem, a non affine inverse problem and a 0D cardiocirculatory model, under the two clinical scenarios of systemic hypertension and ventricular septal defect. For the latter case, we perform forward uncertainty quantification.

math.NA↗

Deformable registration and generative modelling of aortic anatomies by auto-decoders and neural ODEs

This work introduces AD-SVFD, a deep learning model for the deformable registration of vascular shapes to a pre-defined reference and for the generation of synthetic anatomies. AD-SVFD operates by representing each geometry as a weighted point cloud and models ambient space deformations as solutions at unit time of ODEs, whose time-independent right-hand sides are expressed through artificial neural networks. The model parameters are optimized by minimizing the Chamfer Distance between the deformed and reference point clouds, while backward integration of the ODE defines the inverse transformation. A distinctive feature of AD-SVFD is its auto-decoder structure, that enables generalization across shape cohorts and favors efficient weight sharing. In particular, each anatomy is associated with a low-dimensional code that acts as a self-conditioning field and that is jointly optimized with the network parameters during training. At inference, only the latent codes are fine-tuned, substantially reducing computational overheads. Furthermore, the use of implicit shape representations enables generative applications: new anatomies can be synthesized by suitably sampling from the latent space and applying the corresponding inverse transformations to the reference geometry. Numerical experiments, conducted on healthy aortic anatomies, showcase the high-quality results of AD-SVFD, which yields extremely accurate approximations at competitive computational costs.

cs.CV↗

Emergence of Structure in Ensembles of Random Neural Networks

Randomness is ubiquitous in many applications across data science and machine learning. Remarkably, systems composed of random components often display emergent global behaviors that appear deterministic, manifesting a transition from microscopic disorder to macroscopic organization. In this work, we introduce a theoretical model for studying the emergence of collective behaviors in ensembles of random classifiers. We argue that, if the ensemble is weighted through the Gibbs measure defined by adopting the classification loss as an energy, then there exists a finite temperature parameter for the distribution such that the classification is optimal, with respect to the loss (or the energy). Interestingly, for the case in which samples are generated by a Gaussian distribution and labels are constructed by employing a teacher perceptron, we analytically prove and numerically confirm that such optimal temperature does not depend neither on the teacher classifier (which is, by construction of the learning problem, unknown), nor on the number of random classifiers, highlighting the universal nature of the observed behavior. Experiments on the MNIST dataset underline the relevance of this phenomenon in high-quality, noiseless, datasets. Finally, a physical analogy allows us to shed light on the self-organizing nature of the studied phenomenon.

cs.LG↗

Physics-informed neural network estimation of active material properties in time-dependent cardiac biomechanical models

Active stress models in cardiac biomechanics account for the mechanical deformation caused by muscle activity, thus providing a link between the electrophysiological and mechanical properties of the tissue. The accurate assessment of active stress parameters is fundamental for a precise understanding of myocardial function but remains difficult to achieve in a clinical setting, especially when only displacement and strain data from medical imaging modalities are available. This work investigates, through an in-silico study, the application of physics-informed neural networks (PINNs) for inferring active contractility parameters in time-dependent cardiac biomechanical models from these types of imaging data. In particular, by parametrising the sought state and parameter field with two neural networks, respectively, and formulating an energy minimisation problem to search for the optimal network parameters, we are able to reconstruct in various settings active stress fields in the presence of noise and with a high spatial resolution. To this end, we also advance the vanilla PINN learning algorithm with the use of adaptive weighting schemes, ad-hoc regularisation strategies, Fourier features, and suitable network architectures. In addition, we thoroughly analyse the influence of the loss weights in the reconstruction of active stress parameters. Finally, we apply the method to the characterisation of tissue inhomogeneities and detection of fibrotic scars in myocardial tissue. This approach opens a new pathway to significantly improve the diagnosis, treatment planning, and management of heart conditions associated with cardiac fibrosis.

cs.LG↗