Search

Class imbalance in binary classification tasks remains a significant challenge in machine learning, often resulting in poor performance on minority classes. This study comprehensively evaluates three widely-used strategies for handling class imbalance: Synthetic Minority Over-sampling Technique (SMOTE), Class Weights tuning, and Decision Threshold Calibration. We compare these methods against a baseline scenario of no-intervention across 15 diverse machine learning models and 30 datasets from various domains, conducting a total of 9,000 experiments. Performance was primarily assessed using the F1-score, although our study also tracked results on additional 9 metrics including F2-score, precision, recall, Brier-score, PR-AUC, and AUC. Our results indicate that all three strategies generally outperform the baseline, with Decision Threshold Calibration emerging as the most consistently effective technique. However, we observed substantial variability in the best-performing method across datasets, highlighting the importance of testing multiple approaches for specific problems. This study provides valuable insights for practitioners dealing with imbalanced datasets and emphasizes the need for dataset-specific analysis in evaluating class imbalance handling techniques., Comment: 13 pages including appendix, 4 tables

In the realm of computer vision, the perception and reconstruction of the 3D world through vision signals heavily rely on camera intrinsic parameters, which have long been a subject of intense research within the community. In practical applications, without a strong scene geometry prior like the Manhattan World assumption or special artificial calibration patterns, monocular focal length estimation becomes a challenging task. In this paper, we propose a method for monocular focal length estimation using category-level object priors. Based on two well-studied existing tasks: monocular depth estimation and category-level object canonical representation learning, our focal solver takes depth priors and object shape priors from images containing objects and estimates the focal length from triplets of correspondences in closed form. Our experiments on simulated and real world data demonstrate that the proposed method outperforms the current state-of-the-art, offering a promising solution to the long-standing monocular focal length estimation problem.

We develop R2N, a modified quasi-Newton method for minimizing the sum of a $\mathcal{C}^1$ function $f$ and a lower semi-continuous prox-bounded $h$. Both $f$ and $h$ may be nonconvex. At each iteration, our method computes a step by minimizing the sum of a quadratic model of $f$, a model of $h$, and an adaptive quadratic regularization term. A step may be computed by a variant of the proximal-gradient method. An advantage of R2N over trust-region (TR) methods is that proximal operators do not involve an extra TR indicator. We also develop the variant R2DH, in which the model Hessian is diagonal, which allows us to compute a step without relying on a subproblem solver when $h$ is separable. R2DH can be used as standalone solver, but also as subproblem solver inside R2N. We describe non-monotone variants of both R2N and R2DH. Global convergence of a first-order stationarity measure to zero holds without relying on local Lipschitz continuity of $\nabla f$, while allowing model Hessians to grow unbounded, an assumption particularly relevant to quasi-Newton models. Under Lipschitz-continuity of $\nabla f$, we establish a tight worst-case complexity bound of $O(1 / \epsilon^{2/(1 - p)})$ to bring said measure below $\epsilon > 0$, where $0 \leq p < 1$ controls the growth of model Hessians. The latter must not diverge faster than $|\mathcal{S}_k|^p$, where $\mathcal{S}_k$ is the set of successful iterations up to iteration $k$. When $p = 1$, we establish the tight exponential complexity bound $O(\exp(c \epsilon^{-2}))$ where $c > 0$ is a constant. We describe our Julia implementation and report numerical experience on a basis-pursuit problem, image denoising, minimum-rank matrix completion, and a nonlinear support vector machine. In particular, the minimum-rank problem cannot be solved directly at this time by a TR approach as corresponding proximal operators are not known analytically.

Text-to-image (T2I) models have significantly advanced the development of artificial intelligence, enabling the generation of high-quality images in diverse contexts based on specific text prompts. However, existing T2I-based methods often struggle to accurately reproduce the appearance of individuals from a reference image and to create novel representations of those individuals in various settings. To address this, we leverage the pre-trained UNet from Stable Diffusion to incorporate the target face image directly into the generation process. Our approach diverges from prior methods that depend on fixed encoders or static face embeddings, which often fail to bridge encoding gaps. Instead, we capitalize on UNet's sophisticated encoding capabilities to process reference images across multiple scales. By innovatively altering the cross-attention layers of the UNet, we effectively fuse individual identities into the generative process. This strategic integration of facial features across various scales not only enhances the robustness and consistency of the generated images but also facilitates efficient multi-reference and multi-identity generation. Our method sets a new benchmark in identity-preserving image generation, delivering state-of-the-art results in similarity metrics while maintaining prompt alignment.

Modern multicore System-on-Chips (SoCs) feature hardware monitoring mechanisms that measure total power consumption. However, these aggregate measurements are often insufficient for fine-grained thermal and power management. This paper presents an enhanced Clustering Blind Power Identification (ICBPI) approach, designed to improve the sensitivity and robustness of the traditional Blind Power Identification (BPI) method. BPI estimates the power consumption of individual cores and models the thermal behavior of an SoC using only thermal sensor data and total power measurements. The proposed ICBPI approach refines BPI's initialization process, particularly improving the non-negative matrix factorization (NNMF) step, which is critical to the accuracy of BPI. ICBPI introduces density-based spatial clustering of applications with noise (DBSCAN) to better align temperature and power consumption data, thereby providing more accurate power consumption estimates. We validate the ICBPI method through two key tasks. The first task evaluates power estimation accuracy across four different multicore architectures, including a heterogeneous processor. Results show that ICBPI significantly enhances accuracy, reducing error rates by 77.56% compared to the original BPI and by 68.44% compared to the state-of-the-art BPISS method. The second task focuses on improving the detection and localization of malicious thermal sensor attacks in heterogeneous processors. The results demonstrate that ICBPI enhances the security and robustness of multicore SoCs against such attacks.

Smartphone-based contactless fingerphoto authentication has become a reliable alternative to traditional contact-based fingerprint biometric systems owing to rapid advances in smartphone camera technology. Despite its convenience, fingerprint authentication through fingerphotos is more vulnerable to presentation attacks, which has motivated recent research efforts towards developing fingerphoto Presentation Attack Detection (PAD) techniques. However, prior PAD approaches utilized supervised learning methods that require labeled training data for both bona fide and attack samples. This can suffer from two key issues, namely (i) generalization:the detection of novel presentation attack instruments (PAIs) unseen in the training data, and (ii) scalability:the collection of a large dataset of attack samples using different PAIs. To address these challenges, we propose a novel unsupervised approach based on a state-of-the-art deep-learning-based diffusion model, the Denoising Diffusion Probabilistic Model (DDPM), which is trained solely on bona fide samples. The proposed approach detects Presentation Attacks (PA) by calculating the reconstruction similarity between the input and output pairs of the DDPM. We present extensive experiments across three PAI datasets to test the accuracy and generalization capability of our approach. The results show that the proposed DDPM-based PAD method achieves significantly better detection error rates on several PAI classes compared to other baseline unsupervised approaches., Comment: Accepted by IJCB 2024

Inspired by the legal doctrine of stare decisis, which leverages precedents (prior cases) for informed decision-making, we explore methods to integrate them into LJP models. To facilitate precedent retrieval, we train a retriever with a fine-grained relevance signal based on the overlap ratio of alleged articles between cases. We investigate two strategies to integrate precedents: direct incorporation at inference via label interpolation based on case proximity and during training via a precedent fusion module using a stacked-cross attention model. We employ joint training of the retriever and LJP models to address latent space divergence between them. Our experiments on LJP tasks from the ECHR jurisdiction reveal that integrating precedents during training coupled with joint training of the retriever and LJP model, outperforms models without precedents or with precedents incorporated only at inference, particularly benefiting sparser articles., Comment: Accepted to EMNLP Findings

Self-supervised pre-training has proven highly effective for many computer vision tasks, particularly when labelled data are scarce. In the context of Earth Observation (EO), foundation models and various other Vision Transformer (ViT)-based approaches have been successfully applied for transfer learning to downstream tasks. However, it remains unclear under which conditions pre-trained models offer significant advantages over training from scratch. In this study, we investigate the effectiveness of pre-training ViT-based Masked Autoencoders (MAE) for downstream EO tasks, focusing on reconstruction, segmentation, and classification. We consider two large ViT-based MAE pre-trained models: a foundation model (Prithvi) and SatMAE. We evaluate Prithvi on reconstruction and segmentation-based downstream tasks, and for SatMAE we assess its performance on a classification downstream task. Our findings suggest that pre-training is particularly beneficial when the fine-tuning task closely resembles the pre-training task, e.g. reconstruction. In contrast, for tasks such as segmentation or classification, training from scratch with specific hyperparameter adjustments proved to be equally or more effective.

We suggest a resonant $c \bar b \to H^+$ production search, followed by bosonic $H^+ \to W^+ H$ weak decay at the Large Hadron Collider (LHC). In the general two Higgs doublet model (G2HDM) that has flavor-changing neutral Higgs couplings, $H^+$ is resonantly produced via the top-charm $\rho_{tc} V_{tb}$ coupling at tree level, while $H^+ \to W^+ H$ weak decay occurs within the exotic second doublet, leading eventually to same sign dilepton signals. We perform a signal-to-background analysis at the 14 TeV LHC and show that discovery seems possible with LHC Run 2 data already at hand., Comment: 5 pages, 3 figures; v2: title, abstract and introduction updated

This paper analyzes a downlink C-NOMA scenario where a base station (BS) is deployed to serve a pair of users equipped with movable antenna (MA) technology. The user with better channel conditions with the BS will be able to transmit the signal to the other user providing an extra transmission resource and enhancing performance. Both users are equipped with a receiving MA each and a transmitting MA for the relaying user. In this regard, we formulate an optimization problem with the objective of maximizing the achievable sum rate by jointly determining the beamforming vector at the BS, the transmit power at the device and the positions of the MAs while meeting the quality of service (QoS) constraints. Due to the non-convex structure of the formulated problem and the randomness in the channels we adopt a deep deterministic policy gradient (DDPG) approach, a reinforcement learning (RL) algorithm capable of dealing with continuous state and action spaces. Numerical results demonstrate the superiority of the presented model compared to the other benchmark schemes showing gains reaching 45% compared to the NOMA enabled MA scheme and 60% compared to C-NOMA model with fixed antennas. The solution approach showed 93% accuracy compared to the optimal solution.

We introduce Atlas-Chat, the first-ever collection of LLMs specifically developed for dialectal Arabic. Focusing on Moroccan Arabic, also known as Darija, we construct our instruction dataset by consolidating existing Darija language resources, creating novel datasets both manually and synthetically, and translating English instructions with stringent quality control. Atlas-Chat-2B, 9B, and 27B models, fine-tuned on the dataset, exhibit superior ability in following Darija instructions and performing standard NLP tasks. Notably, our models outperform both state-of-the-art and Arabic-specialized LLMs like LLaMa, Jais, and AceGPT, e.g., our 9B model gains a 13% performance boost over a larger 13B model on DarijaMMLU, in our newly introduced evaluation suite for Darija covering both discriminative and generative tasks. Furthermore, we perform an experimental analysis of various fine-tuning strategies and base model choices to determine optimal configurations. All our resources are publicly accessible, and we believe our work offers comprehensive design methodologies of instruction-tuning for low-resource languages, which are often neglected in favor of data-rich languages by contemporary LLMs.

Signal Processing (SP) and Machine Learning (ML) rely on good math and coding knowledge, in particular, linear algebra, probability, and complex numbers. A good grasp of these relies on scalar algebra learned in middle school. The ability to understand and use scalar algebra well, in turn, relies on a good foundation in basic arithmetic. Because of various systemic barriers, many students are not able to build a strong foundation in arithmetic in elementary school. This leads them to struggle with algebra and everything after that. Since math learning is cumulative, the gap between those without a strong early foundation and everyone else keeps increasing over the school years and becomes difficult to fill in college. In this article we discuss how SP faculty and graduate students can play an important role in starting, and participating in, university-run (or other) out-of-school math support programs to supplement students' learning. Two example programs run by the authors (CyMath at ISU and Ab7G at Purdue) are briefly described. The second goal of this article is to use our perspective as SP, and engineering, educators who have seen the long-term impact of elementary school math teaching policies, to provide some simple almost zero cost suggestions that elementary schools could adopt to improve math learning: (i) more math practice in school, (ii) send small amounts of homework (individual work is critical in math), and (iii) parent awareness (math resources, need for early math foundation, clear in-school test information and sharing of feedback from the tests). In summary, good early math support (in school and through out-of-school programs) can help make SP and ML more accessible., Comment: Under submission to IEEE Signal Processing Magazine

This paper investigates the structural functional observability (SFO) and structural output controllability (SOC) of a class of systems with generically diagonalizable state matrices and explores the associated minimal sensor and actuator placement problems. The verification of SOC and the corresponding sensor and actuator placement problems, i.e., the problems of determining the minimum number of outputs and inputs required to achieve SFO and SOC, respectively, are yet open for general systems, which motivates our focus on a class of systems enabling polynomial-time solutions. In this line, we first define and characterize generically diagonalizable systems, referring to structured systems for which almost all realizations of the state matrices are diagonalizable. We then develop computationally efficient criteria for SFO and SOC within the context of generically diagonalizable systems. Our work expands the class of systems amenable to polynomial-time SOC verification. Thanks to the simplicity of the obtained criteria, we derive closed-form solutions for determining the minimal sensor placement to achieve SFO and the minimal actuator deployment to achieve SOC in such systems, along with efficient weighted maximum matching based and weighted maximum flow based algorithms. For more general systems to achieve SFO, an upper bound is given by identifying a non-decreasing property of SFO with respect to a specific class of edge additions, which is shown to be optimal under certain circumstances., Comment: Under review in a Journal

Elastoviscoplastic (EVP) fluids, which exhibit both solid-like and liquid-like behavior depending on the applied stress, are critical in industrial processes involving complex geometries such as porous media and wavy channels. In this study, we investigate how flow characteristics and channel design affect EVP fluid flow through a wavy channel, using numerical simulations supported by microfluidic experiments. Our results reveal that elasticity significantly influences flow dynamics, reducing pressure drops and expanding unyielded regions. Notably, we find that even minimal elasticity can shift the flow from steady to time-dependent regimes, a transition less pronounced in viscoelastic fluids. Additionally, we show that the development of stagnation regions can be prevented when using a modified EVP fluid with enhanced elasticity, thus providing a full global yielding of the material. This study elucidates the role of elasticity in modifying flow patterns and stress distribution within EVP fluids, offering insights into the optimization of industrial applications, such as the displacement of yield stress fluids in enhanced oil recovery, gas extraction, cementing, and other processes where flow efficiency is critical., Comment: 15 pages, 19 figures

This work presents a control-oriented identification scheme for efficient control design and stability analysis of nonlinear systems. Neural networks are used to identify a discrete-time nonlinear state-space model to approximate time-domain input-output behavior of a nonlinear system. The network is constructed such that the identified model is approximately linearizable by feedback, ensuring that the control law trivially follows from the learning stage. After the identification and quasi-linearization procedures, linear control theory comes at hand to design robust controllers and study stability of the closed-loop system. The effectiveness and interest of the methodology are illustrated throughout the paper on popular benchmarks for system identification., Comment: 15 pages, 3 figures, 6 tables, accepted as a poster in SysDO 2024, Stuttgart, Germany

In many critical Machine Learning applications, such as autonomous driving and medical image diagnosis, the detection of out-of-distribution (OOD) samples is as crucial as accurately classifying in-distribution (ID) inputs. Recently Outlier Exposure (OE) based methods have shown promising results in detecting OOD inputs via model fine-tuning with auxiliary outlier data. However, most of the previous OE-based approaches emphasize more on synthesizing extra outlier samples or introducing regularization to diversify OOD sample space, which is rather unquantifiable in practice. In this work, we propose a novel and straightforward method called Margin bounded Confidence Scores (MaCS) to address the nontrivial OOD detection problem by enlarging the disparity between ID and OOD scores, which in turn makes the decision boundary more compact facilitating effective segregation with a simple threshold. Specifically, we augment the learning objective of an OE regularized classifier with a supplementary constraint, which penalizes high confidence scores for OOD inputs compared to that of ID and significantly enhances the OOD detection performance while maintaining the ID classification accuracy. Extensive experiments on various benchmark datasets for image classification tasks demonstrate the effectiveness of the proposed method by significantly outperforming state-of-the-art (S.O.T.A) methods on various benchmarking metrics. The code is publicly available at https://github.com/lakpa-tamang9/margin_ood, Comment: 10 pages, 5 figures, IEEE Conference in Data Mining 2024

Few-shot learning (FSL) aims to enable models to recognize novel objects or classes with limited labelled data. Feature generators, which synthesize new data points to augment limited datasets, have emerged as a promising solution to this challenge. This paper investigates the effectiveness of feature generators in enhancing the embedding process for FSL tasks. To address the issue of inaccurate embeddings due to the scarcity of images per class, we introduce a feature generator that creates visual features from class-level textual descriptions. By training the generator with a combination of classifier loss, discriminator loss, and distance loss between the generated features and true class embeddings, we ensure the generation of accurate same-class features and enhance the overall feature representation. Our results show a significant improvement in accuracy over baseline methods, with our approach outperforming the baseline model by 10% in 1-shot and around 5% in 5-shot approaches. Additionally, both visual-only and visual + textual generators have also been tested in this paper. The code is publicly available at https://github.com/heethanjan/Feature-Generator-for-FSL., Comment: 16 pages, Accepted to ACCV 2024

In this work, we propose a theoretical scheme to explore the enhancement of quantum correlation hierarchies in a cavity magnonmechanical system via the coherent feedback tool. We use Gaussian geometric discord to quantify quantum correlations between the two magnon modes, including those beyond entanglement, in the steady state. Logarithmic negativity and Gaussian quantum steering are employed to characterize entanglement and steerability, respectively. Our results show that adjusting the beam splitter's reflective parameter can significantly enhance quantum correlations and increase their resilience to thermal noise. Moreover, we demonstrate that coherent feedback can achieve enhanced genuine tripartite entanglement among the photon, magnon \(M_1\), and phonon. These findings present promising strategies for enhancing entanglement in magnon-based systems and advancing quantum information technologies. We conclude by validating the system and demonstrating its ability to detect entanglement., Comment: 10 pages, 5 figures

In this article, we are interested in endomorphism's impact on the commutativity of a Banach algebra. Our research uses a topological approach, drawing on specific results from functional analysis and algebraic techniques.\;Also, we provide examples to show that we need certain assumptions.

For any positive integer $n$, we show that there exists a real number field $k$ (resp. $k'$) of degree $2^n$ whose $2$-class group is isomorphic $\mathbb{Z}/2\mathbb{Z}\times \mathbb{Z}/2\mathbb{Z}$ such that the Galois group of the maximal unramified extension of $k$ (resp. $k'$) over $k$ (resp. $k'$) is abelian (resp. non abelian, more precisely isomorphic to $Q_8$ or $D_8$, the quaternion and the dihedral group of order $8$ respectively). In fact, we construct the first examples in literature of families of real biquadratic fields whose unramified abelian Iwasawa module is isomorphic to $\mathbb{Z}/2\mathbb{Z}\times \mathbb{Z}/2\mathbb{Z}$, and so that satisfying the Greenberg conjecture., Comment: 12 pages

This paper addresses the Cauchy problem for wave equations with scale-invariant time-dependent damping and nonlinear time-derivative terms, modeled as $$\partial_{t}^2u- \Delta u +\frac{\mu}{1+t}\partial_tu= f(\partial_tu), \quad x\in \mathbb{R}^n, t>0,$$ where $f(\partial_tu)=|\partial_tu|^p $ or $|\partial_tu|^{p-1}\partial_tu$ with $p>1$ and $\mu>0$. We prove global existence of small data solutions in low dimensions $1\leq n\leq 3$ by using energy estimates in appropriate Sobolev spaces. Our primary contribution is an existence result for $p>1+\frac2{\mu}$, in the one-dimensional case, when $\mu \le 2$, which in conjunction with prior blow-up results from \cite{Our2}, establish that the critical exponent for small data solutions in one dimension is $p_G(1,\mu)=1+\frac2{\mu}$, when $\mu \le 2$. To the best of our knowledge, this is the first identification of the critical exponent range for the time-dependent damped wave equations with scale-invariant and time-derivative nonlinearity., Comment: 22 pages

The remarkable advancements in artificial intelligence (AI), primarily driven by deep neural networks, are facing challenges surrounding unsustainable computational trajectories, limited robustness, and a lack of explainability. To develop next-generation cognitive AI systems, neuro-symbolic AI emerges as a promising paradigm, fusing neural and symbolic approaches to enhance interpretability, robustness, and trustworthiness, while facilitating learning from much less data. Recent neuro-symbolic systems have demonstrated great potential in collaborative human-AI scenarios with reasoning and cognitive capabilities. In this paper, we aim to understand the workload characteristics and potential architectures for neuro-symbolic AI. We first systematically categorize neuro-symbolic AI algorithms, and then experimentally evaluate and analyze them in terms of runtime, memory, computational operators, sparsity, and system characteristics on CPUs, GPUs, and edge SoCs. Our studies reveal that neuro-symbolic models suffer from inefficiencies on off-the-shelf hardware, due to the memory-bound nature of vector-symbolic and logical operations, complex flow control, data dependencies, sparsity variations, and limited scalability. Based on profiling insights, we suggest cross-layer optimization solutions and present a hardware acceleration case study for vector-symbolic architecture to improve the performance, efficiency, and scalability of neuro-symbolic computing. Finally, we discuss the challenges and potential future directions of neuro-symbolic AI from both system and architectural perspectives., Comment: 14 pages, 11 figures, 7 tables; IEEE Transactions on Circuits and Systems for Artificial Intelligence (TCASAI), 2024

Autonomous driving at unsignalized intersections is still considered a challenging application for machine learning due to the complications associated with handling complex multi-agent scenarios characterized by a high degree of uncertainty. Automating the decision-making process at these safety-critical environments involves comprehending multiple levels of abstractions associated with learning robust driving behaviors to enable the vehicle to navigate efficiently. In this survey, we aim at exploring the state-of-the-art techniques implemented for decision-making applications, with a focus on algorithms that combine Reinforcement Learning (RL) and deep learning for learning traversing policies at unsignalized intersections. The reviewed schemes vary in the proposed driving scenario, in the assumptions made for the used intersection model, in the tackled challenges, and in the learning algorithms that are used. We have presented comparisons for these techniques to highlight their limitations and strengths. Based on our in-depth investigation, it can be discerned that a robust decision-making scheme for navigating real-world unsignalized intersection has yet to be developed. Along with our analysis and discussion, we recommend potential research directions encouraging the interested players to tackle the highlighted challenges. By adhering to our recommendations, decision-making architectures that are both non-overcautious and safe, yet feasible, can be trained and validated in real-world unsignalized intersections environments.

In this paper, a space-air-ground quantum (SPARQ) network is developed as a means for providing a seamless on-demand entanglement distribution. The node mobility in SPARQ poses significant challenges to entanglement routing. Existing quantum routing algorithms focus on stationary ground nodes and utilize link distance as an optimality metric, which is unrealistic for dynamic systems like SPARQ. Moreover, in contrast to the prior art that assumes homogeneous nodes, SPARQ encompasses heterogeneous nodes with different functionalities further complicates the entanglement distribution. To solve the entanglement routing problem, a deep reinforcement learning (RL) framework is proposed and trained using deep Q-network (DQN) on multiple graphs of SPARQ to account for the network dynamics. Subsequently, an entanglement distribution policy, third-party entanglement distribution (TPED), is proposed to establish entanglement between communication parties. A realistic quantum network simulator is designed for performance evaluation. Simulation results show that the TPED policy improves entanglement fidelity by 3% and reduces memory consumption by 50% compared with benchmark. The results also show that the proposed DQN algorithm improves the number of resolved teleportation requests by 39% compared with shortest path baseline and the entanglement fidelity by 2% compared with an RL algorithm that is based on long short-term memory (LSTM). It also improved entanglement fidelity by 6% and 9% compared with two state-of-the-art benchmarks. Moreover, the entanglement fidelity is improved by 15% compared with DQN trained on a snapshot of SPARQ. Additionally, SPARQ enhances the average entanglement fidelity by 23.5% compared with existing networks spanning only space and ground layers.

This research presents a comprehensive framework for transitioning financial diffusion models from the risk-neutral (RN) measure to the real-world (RW) measure, leveraging results from probability theory, specifically Girsanov's theorem. The RN measure, fundamental in derivative pricing, is contrasted with the RW measure, which incorporates risk premiums and better reflects actual market behavior and investor preferences, making it crucial for risk management. We address the challenges of incorporating real-world dynamics into financial models, such as accounting for market premiums, producing realistic term structures of market indicators, and fitting any arbitrarily given market curve. Our framework is designed to be general, applicable to a variety of diffusion models, including those with non-additive noise such as the CIR++ model. Through case studies involving Goldman Sachs' 2024 global credit outlook forecasts and the European Banking Authority (EBA) 2023 stress tests, we validate the robustness, practical relevance and applicability of our methodology. This work contributes to the literature by providing a versatile tool for better risk measures and enhancing the realism of financial models under the RW measure. Our model's versatility extends to stress testing and scenario analysis, providing practitioners with a powerful tool to evaluate various what-if scenarios and make well-informed decisions, particularly in pricing and risk management strategies.

In this article we consider the surplus process of an insurance company within the CramerLundberg framework. We study the optimal reinsurance strategy and dividend distribution of an insurance company under proportional reinsurance, in which capital injections are allowed. Our aim is to find a general dynamic reinsurance strategy that maximizes the expected discounted cumulative dividends until the time of passage below a given level, called ruin. These policies consist in stopping at the first time when the size of the overshoot below 0 exceeds a certain limit, and only pay dividends when the reserve reaches an upper barrier. Using analytical methods, we identify the value function as a particular solution to the associated Hamilton Jacobi Bellman equation. This approach leads to an exhaustive and explicit characterization of optimal policy. The proportional reinsurance is given via comprehensive structure equations. Furthermore we give some examples illustrating the applicability of this method for proportional reinsurance treaties.

We present AutoMode-ASR, a novel framework that effectively integrates multiple ASR systems to enhance the overall transcription quality while optimizing cost. The idea is to train a decision model to select the optimal ASR system for each segment based solely on the audio input before running the systems. We achieve this by ensembling binary classifiers determining the preference between two systems. These classifiers are equipped with various features, such as audio embeddings, quality estimation, and signal properties. Additionally, we demonstrate how using a quality estimator can further improve performance with minimal cost increase. Experimental results show a relative reduction in WER of 16.2%, a cost saving of 65%, and a speed improvement of 75%, compared to using a single-best model for all segments. Our framework is compatible with commercial and open-source black-box ASR systems as it does not require changes in model codes., Comment: SPECOM 2024 Conference

Real-time rendering of human head avatars is a cornerstone of many computer graphics applications, such as augmented reality, video games, and films, to name a few. Recent approaches address this challenge with computationally efficient geometry primitives in a carefully calibrated multi-view setup. Albeit producing photorealistic head renderings, it often fails to represent complex motion changes such as the mouth interior and strongly varying head poses. We propose a new method to generate highly dynamic and deformable human head avatars from multi-view imagery in real-time. At the core of our method is a hierarchical representation of head models that allows to capture the complex dynamics of facial expressions and head movements. First, with rich facial features extracted from raw input frames, we learn to deform the coarse facial geometry of the template mesh. We then initialize 3D Gaussians on the deformed surface and refine their positions in a fine step. We train this coarse-to-fine facial avatar model along with the head pose as a learnable parameter in an end-to-end framework. This enables not only controllable facial animation via video inputs, but also high-fidelity novel view synthesis of challenging facial expressions, such as tongue deformations and fine-grained teeth structure under large motion changes. Moreover, it encourages the learned head avatar to generalize towards new facial expressions and head poses at inference time. We demonstrate the performance of our method with comparisons against the related methods on different datasets, spanning challenging facial expression sequences across multiple identities. We also show the potential application of our approach by demonstrating a cross-identity facial performance transfer application., Comment: ACM Transaction on Graphics (SIGGRAPH Asia 2024); Project page: https://vcai.mpi-inf.mpg.de/projects/GaussianHeads/

Accurate recognition of human emotions is a crucial challenge in affective computing and human-robot interaction (HRI). Emotional states play a vital role in shaping behaviors, decisions, and social interactions. However, emotional expressions can be influenced by contextual factors, leading to misinterpretations if context is not considered. Multimodal fusion, combining modalities like facial expressions, speech, and physiological signals, has shown promise in improving affect recognition. This paper proposes a transformer-based multimodal fusion approach that leverages facial thermal data, facial action units, and textual context information for context-aware emotion recognition. We explore modality-specific encoders to learn tailored representations, which are then fused using additive fusion and processed by a shared transformer encoder to capture temporal dependencies and interactions. The proposed method is evaluated on a dataset collected from participants engaged in a tangible tabletop Pacman game designed to induce various affective states. Our results demonstrate the effectiveness of incorporating contextual information and multimodal fusion for affective state recognition.

In material physics, characterization techniques are foremost crucial for obtaining the materials data regarding the physical properties as well as structural, electronics, magnetic, optic, dielectric, and spectroscopic characteristics. However, for many materials, ensuring availability and safe accessibility is not always easy and fully warranted. Moreover, the use of modeling and simulation techniques need a lot of theoretical knowledge, in addition of being associated to costly computation time and a great complexity deal. Thus, analyzing materials with different techniques for multiple samples simultaneously, still be very challenging for engineers and researchers. It is worth noting that although of being very risky, X-ray diffraction is the well known and widely used characterization technique which gathers data from structural properties of crystalline 1d, 2d or 3d materials. We propose in this paper, a Smart GRU for Gated Recurrent Unit model to forcast structural characteristics or properties of thin films of tin oxide SnO$_2$(110). Indeed, thin films samples are elaborated and managed experimentally and the collected data dictionary is then used to generate an AI -- Artificial Intelligence -- GRU model for the thin films of tin oxide SnO$_2$(110) structural property characterization., Comment: 19 pages, 14 figures. Baltica Journal, Special Issues, September 2024

Traffic management and on-road safety have been a concern for the transportation authorities and the engineering communities for many years. Most of the implemented technologies for intelligent highways focus on safety measures and increased driver awareness, and expect a centralized management for the vehicular traffic flow. Leveraging recent advances in wireless communication, researchers have proposed solutions based on vehicle-to-vehicle (V2V) and vehicle-to-Infrastructure (V2I) communication in order to detect traffic jams and better disseminate data from on-road and on-vehicle sensors. Moreover, the development of connected autonomous vehicles (CAV) have motivated a paradigm shift in how traffic will be managed. Overall, these major technological advances have motivated the notion of dynamic traffic management (DTM), where smart road reconfiguration capabilities, e.g., dynamic lane reversal, adaptive traffic light timing, etc. will be exploited in real-time to improve traffic flow and adapt to unexpected incidents. This chapter discusses what the challenges in realizing DTM are and covers how CAV has revolutionized traffic management. Moreover, we highlight the issues for handling human-driven vehicles while roads are transitioning to CAV only traffic. Particularly, we articulate a new vision for inter-vehicle communication and assessment of road conditions, and promote a novel system for traffic management. Vehicle to on-road sensors as well as inter-vehicle connectivity will be enabled through the use of handheld devices such as smartphones. This not only enables real-time data sharing but also expedites the adoption of DTM without awaiting the dominant presence of autonomous vehicle on the road. ...

The major advances in intelligent transportation systems are pushing societal services toward autonomy where road management is to be more agile in order to cope with changes and continue to yield optimal performance. However, the pedestrian experience is not sufficiently considered. Particularly, signalized intersections are expected to be popular if not dominant in urban settings where pedestrian density is high. This paper presents the design of a novel environment for simulating human motion on signalized crosswalks at a fine-grained level. Such a simulation not only captures typical behavior, but also handles cases where large pedestrian groups cross from both directions. The proposed simulator is instrumental for optimized road configuration management where the pedestrians' quality of experience, for example, waiting time, is factored in. The validation results using field data show that an accuracy of 98.37 percent can be obtained for the estimated crossing time. Other results using synthetic data show that our simulator enables optimized traffic light scheduling that diminishes pedestrians' waiting time without sacrificing vehicular throughput.

Motivated by structural properties of differential field extensions, we introduce the notion of a theory $T$ being derivation-like with respect to another model-complete theory $T_0$. We prove that when $T$ admits a model-companion $T_+$, then several model-theoretic properties transfer from $T_0$ to $T_+$. These properties include completeness, quantifier-elimination, stability, simplicity, and NSOP$_1$. We also observe that, aside from the theory of differential fields, examples of derivation-like theories are plentiful.

The emerging field of smart agriculture leverages the Internet of Things (IoT) to revolutionize farming practices. This paper investigates the transformative potential of Long Range (LoRa) technology as a key enabler of long-range wireless communication for agricultural IoT systems. By reviewing existing literature, we identify a gap in research specifically focused on LoRa's prospects and challenges from a communication perspective in smart agriculture. We delve into the details of LoRa-based agricultural networks, covering network architecture design, Physical Layer (PHY) considerations tailored to the agricultural environment, and channel modeling techniques that account for soil characteristics. The paper further explores relaying and routing mechanisms that address the challenges of extending network coverage and optimizing data transmission in vast agricultural landscapes. Transitioning to practical aspects, we discuss sensor deployment strategies and energy management techniques, offering insights for real-world deployments. A comparative analysis of LoRa with other wireless communication technologies employed in agricultural IoT applications highlights its strengths and weaknesses in this context. Furthermore, the paper outlines several future research directions to leverage the potential of LoRa-based agriculture 4.0. These include advancements in channel modeling for diverse farming environments, novel relay routing algorithms, integrating emerging sensor technologies like hyper-spectral imaging and drone-based sensing, on-device Artificial Intelligence (AI) models, and sustainable solutions. This survey can guide researchers, technologists, and practitioners to understand, implement, and propel smart agriculture initiatives using LoRa technology.

Federated Learning (FL) has gained attention across various industries for its capability to train machine learning models without centralizing sensitive data. While this approach offers significant benefits such as privacy preservation and decreased communication overhead, it presents several challenges, including deployment complexity and interoperability issues, particularly in heterogeneous scenarios or resource-constrained environments. Over-the-air (OTA) FL was introduced to tackle these challenges by disseminating model updates without necessitating direct device-to-device connections or centralized servers. However, OTA-FL brought forth limitations associated with heightened energy consumption and network latency. In this paper, we propose a multi-attribute client selection framework employing the grey wolf optimizer (GWO) to strategically control the number of participants in each round and optimize the OTA-FL process while considering accuracy, energy, delay, reliability, and fairness constraints of participating devices. We evaluate the performance of our multi-attribute client selection approach in terms of model loss minimization, convergence time reduction, and energy efficiency. In our experimental evaluation, we assessed and compared the performance of our approach against the existing state-of-the-art methods. Our results demonstrate that the proposed GWO-based client selection outperforms these baselines across various metrics. Specifically, our approach achieves a notable reduction in model loss, accelerates convergence time, and enhances energy efficiency while maintaining high fairness and reliability indicators.

Analyzing and certifying stability and attractivity of nonlinear systems is a topic of research interest that has been extensively investigated by control theorists and engineers for many years. Despite that, accurately estimating domains of attraction for nonlinear systems remains a challenging task, where available estimation approaches are either conservative or limited to low-dimensional systems. In this work, we propose an iterative approach to accurately underapproximate safe (i.e., state-constrained) domains of attraction for general discrete-time autonomous nonlinear systems. Our approach relies on implicit representations of safe backward reachable sets of safe regions of attraction, where such regions can be be easily constructed using, e.g., quadratic Lyapunov functions. The iterations of our approach are monotonic (in the sense of set inclusion), where each iteration results in a safe region of attraction, given as a sublevel set, that underapproximates the safe domain of attraction. The sublevel set representations of the resulting regions of attraction can be efficiently utilized in verifying the inclusion of given points of interest in the safe domain of attraction. We illustrate our approach through two numerical examples, involving two- and four-dimensional nonlinear systems.

The performance of diagnostic Computer-Aided Design (CAD) systems for retinal diseases depends on the quality of the retinal images being screened. Thus, many studies have been developed to evaluate and assess the quality of such retinal images. However, most of them did not investigate the relationship between the accuracy of the developed models and the quality of the visualization of interpretability methods for distinguishing between gradable and non-gradable retinal images. Consequently, this paper presents a novel framework called FGR-Net to automatically assess and interpret underlying fundus image quality by merging an autoencoder network with a classifier network. The FGR-Net model also provides an interpretable quality assessment through visualizations. In particular, FGR-Net uses a deep autoencoder to reconstruct the input image in order to extract the visual characteristics of the input fundus images based on self-supervised learning. The extracted features by the autoencoder are then fed into a deep classifier network to distinguish between gradable and ungradable fundus images. FGR-Net is evaluated with different interpretability methods, which indicates that the autoencoder is a key factor in forcing the classifier to focus on the relevant structures of the fundus images, such as the fovea, optic disk, and prominent blood vessels. Additionally, the interpretability methods can provide visual feedback for ophthalmologists to understand how our model evaluates the quality of fundus images. The experimental results showed the superiority of FGR-Net over the state-of-the-art quality assessment methods, with an accuracy of 89% and an F1-score of 87%.

This paper presents an explainable machine learning (ML) approach for predicting surface roughness in milling. Utilizing a dataset from milling aluminum alloy 2017A, the study employs random forest regression models and feature importance techniques. The key contributions include developing ML models that accurately predict various roughness values and identifying redundant sensors, particularly those for measuring normal cutting force. Our experiments show that removing certain sensors can reduce costs without sacrificing predictive accuracy, highlighting the potential of explainable machine learning to improve cost-effectiveness in machining., Comment: arXiv admin note: substantial text overlap with arXiv:2403.18731

This study compares the efficacy of GPT-4 and clinalytix Medical AI in predicting the clinical risk of delirium development. Findings indicate that GPT-4 exhibited significant deficiencies in identifying positive cases and struggled to provide reliable probability estimates for delirium risk, while clinalytix Medical AI demonstrated superior accuracy. A thorough analysis of the large language model's (LLM) outputs elucidated potential causes for these discrepancies, consistent with limitations reported in extant literature. These results underscore the challenges LLMs face in accurately diagnosing conditions and interpreting complex clinical data. While LLMs hold substantial potential in healthcare, they are currently unsuitable for independent clinical decision-making. Instead, they should be employed in assistive roles, complementing clinical expertise. Continued human oversight remains essential to ensure optimal outcomes for both patients and healthcare providers.

Purpose: To develop and evaluate an automated system for extracting structured clinical information from unstructured radiology and pathology reports using open-weights large language models (LMs) and retrieval augmented generation (RAG), and to assess the effects of model configuration variables on extraction performance. Methods and Materials: The study utilized two datasets: 7,294 radiology reports annotated for Brain Tumor Reporting and Data System (BT-RADS) scores and 2,154 pathology reports annotated for isocitrate dehydrogenase (IDH) mutation status. An automated pipeline was developed to benchmark the performance of various LMs and RAG configurations. The impact of model size, quantization, prompting strategies, output formatting, and inference parameters was systematically evaluated. Results: The best performing models achieved over 98% accuracy in extracting BT-RADS scores from radiology reports and over 90% for IDH mutation status extraction from pathology reports. The top model being medical fine-tuned llama3. Larger, newer, and domain fine-tuned models consistently outperformed older and smaller models. Model quantization had minimal impact on performance. Few-shot prompting significantly improved accuracy. RAG improved performance for complex pathology reports but not for shorter radiology reports. Conclusions: Open LMs demonstrate significant potential for automated extraction of structured clinical data from unstructured clinical reports with local privacy-preserving application. Careful model selection, prompt engineering, and semi-automated optimization using annotated data are critical for optimal performance. These approaches could be reliable enough for practical use in research workflows, highlighting the potential for human-machine collaboration in healthcare data extraction.

This paper finds all Lucas numbers which are the sum of two Jacobsthal numbers. It also finds all Jacobsthal numbers which are the sum of two Lucas numbers. In general, we find all non-negative integer solutions $(n, m, k)$ of the two Diophantine equations $L_n +L_m =J_k$ and $J_n +J_m =L_K,$ where $\left\lbrace L_{k}\right\rbrace_{k\geq0}$ and $\left\lbrace J_{n}\right\rbrace_{n\geq0}$ are the sequences of Lucas and Jacobsthal numbers, respectively. Our primary results are supported by an adaption of the Baker's theorem for linear forms in logarithms and Dujella and Peth\H{o}'s reduction method.

Vision-language models have recently evolved into versatile systems capable of high performance across a range of tasks, such as document understanding, visual question answering, and grounding, often in zero-shot settings. Comics Understanding, a complex and multifaceted field, stands to greatly benefit from these advances. Comics, as a medium, combine rich visual and textual narratives, challenging AI models with tasks that span image classification, object detection, instance segmentation, and deeper narrative comprehension through sequential panels. However, the unique structure of comics -- characterized by creative variations in style, reading order, and non-linear storytelling -- presents a set of challenges distinct from those in other visual-language domains. In this survey, we present a comprehensive review of Comics Understanding from both dataset and task perspectives. Our contributions are fivefold: (1) We analyze the structure of the comics medium, detailing its distinctive compositional elements; (2) We survey the widely used datasets and tasks in comics research, emphasizing their role in advancing the field; (3) We introduce the Layer of Comics Understanding (LoCU) framework, a novel taxonomy that redefines vision-language tasks within comics and lays the foundation for future work; (4) We provide a detailed review and categorization of existing methods following the LoCU framework; (5) Finally, we highlight current research challenges and propose directions for future exploration, particularly in the context of vision-language models applied to comics. This survey is the first to propose a task-oriented framework for comics intelligence and aims to guide future research by addressing critical gaps in data availability and task definition. A project associated with this survey is available at https://github.com/emanuelevivoli/awesome-comics-understanding., Comment: under review. project website: https://github.com/emanuelevivoli/awesome-comics-understanding

This paper introduces a novel stochastic model for credit spreads. The stochastic approach leverages the diffusion of default intensities via a CIR++ model and is formulated within a risk-neutral probability space. Our research primarily addresses two gaps in the literature. The first is the lack of credit spread models founded on a stochastic basis that enables continuous modeling, as many existing models rely on factorial assumptions. The second is the limited availability of models that directly yield a term structure of credit spreads. An intermediate result of our model is the provision of a term structure for the prices of defaultable bonds. We present the model alongside an innovative, practical, and conservative calibration approach that minimizes the error between historical and theoretical volatilities of default intensities. We demonstrate the robustness of both the model and its calibration process by comparing its behavior to historical credit spread values. Our findings indicate that the model not only produces realistic credit spread term structure curves but also exhibits consistent diffusion over time. Additionally, the model accurately fits the initial term structure of implied survival probabilities and provides an analytical expression for the credit spread of any given maturity at any future time.

Motivated by string theory scenarios, we study the optical aspect of AdS black holes in Einstein-Maxwell-dilaton theory. Concretely, we investigate and examine the shadows and the deflection angle of light rays by such cosmological black holes. Concerning the shadows, we first deal with the non-rotating solutions. In particular, we obtain perfect circular shadows where their sizes are controlled by the involved parameter including the charge and the cosmological constant. Combining the Newman-Janis formalism and the Hamilton-Jacobi algorithm, we approach the rotating black hole solutions using real curves. Among others, we observe that the size and shape shadows depend on the rotating parameter and the remaining ones. To make contact with Event Horizon Telescope observational data, we show that certain constrains should be imposed on such parameters. Then, we study the behaviors of the light rays near to such cosmological black holes by computing the deflection angle in terms of Einstein-Maxwell-dilaton theory parameters. Specifically, we reveal that the effect of the cosmological constant on the deflection angle changes as the coupling between the black hole parameters is adjusted. When the rotation parameter is introduced, we observe this effect becomes similar to that of the cosmological constant in ordinary AdS black holes., Comment: 25 Pages, 6 Figures, 1 Table, Authors in alphabetical order

The ever-increasing demand for higher data rates in communication systems intensifies the need for advanced non-linear equalizers capable of higher performance. Recently artificial neural networks (ANNs) were introduced as a viable candidate for advanced non-linear equalizers, as they outperform traditional methods. However, they are computationally complex and therefore power hungry. Spiking neural networks (SNNs) started to gain attention as an energy-efficient alternative to ANNs. Recent works proved that they can outperform ANNs at this task. In this work, we explore the design space of an SNN-based decision-feedback equalizer (DFE) to reduce its computational complexity for an efficient implementation on field programmable gate array (FPGA). Our Results prove that it achieves higher communication performance than ANN-based DFE at roughly the same throughput and at 25X higher energy efficiency., Comment: accepted for publication in IEEE Middle East Conference on Communications and Networking (MECOM 2024). November, 2024

The automation of guitar tablature generation from video inputs holds significant promise for enhancing music education, transcription accuracy, and performance analysis. Existing methods face challenges with consistency and completeness, particularly in detecting fretboards and accurately identifying notes. To address these issues, this paper introduces an advanced approach leveraging deep learning, specifically YOLO models for real-time fretboard detection, and Fourier Transform-based audio analysis for precise note identification. Experimental results demonstrate substantial improvements in detection accuracy and robustness compared to traditional techniques. This paper outlines the development, implementation, and evaluation of these methodologies, aiming to revolutionize guitar instruction by automating the creation of guitar tabs from video recordings.

We investigate the initial value problem for a defocusing semi-linear wave equation with spatially growing nonlinearity. Our analysis leads to global well-posedness in the energy space. Furthermore, we obtain the linearization of energy-bounded solutions using the methodology outlined in [8]. The proof hinges on Moser-Trudinger type inequalities and Strichartz estimates., Comment: Typos corrected and text revised

This paper presents a comprehensive study of top quark phenomenology within the Non-Commutative Standard Model. We calculate non-commutative corrections to the squared amplitudes for top quark pair production, as well as for single top quark production via the $t$-channel and $tW$-channel, while noting that the $s$-channel remains unaffected by non-commutative geometry. Utilizing MadGraph5_aMC@NLO, we determine total cross-sections at various center-of-mass energies and examine differential distributions in transverse momentum and rapidity at leading order in both the strong coupling $\alpha_S$ and the non-commutative parameter $\Theta$. For single top production in the $t$-channel, a matching technique is employed to extend the validity of the distribution to low $p_t^{top}$ values through resummation. We also compare the non-commutative corrections with higher-order QCD corrections obtained at NLO using MCFM and at NNLO from existing literature. Our findings reveal significant deviations arising from non-commutative geometry at high energies, providing insights into potential new physics at energy scales accessible by current and future colliders., Comment: 15 pages, 12 figures

We demonstrate that preference optimization methods can effectively enhance LLM safety. Applying various alignment techniques to the Falcon 11B model using safety datasets, we achieve a significant boost in global safety score (from $57.64\%$ to $99.90\%$) as measured by LlamaGuard 3 8B, competing with state-of-the-art models. On toxicity benchmarks, average scores in adversarial settings dropped from over $0.6$ to less than $0.07$. However, this safety improvement comes at the cost of reduced general capabilities, particularly in math, suggesting a trade-off. We identify noise contrastive alignment (Safe-NCA) as an optimal method for balancing safety and performance. Our study ultimately shows that alignment techniques can be sufficient for building safe and robust models.

Autonomous robots use simultaneous localization and mapping (SLAM) for efficient and safe navigation in various environments. LiDAR sensors are integral in these systems for object identification and localization. However, LiDAR systems though effective in detecting solid objects (e.g., trash bin, bottle, etc.), encounter limitations in identifying semitransparent or non-tangible objects (e.g., fire, smoke, steam, etc.) due to poor reflecting characteristics. Additionally, LiDAR also fails to detect features such as navigation signs and often struggles to detect certain hazardous materials that lack a distinct surface for effective laser reflection. In this paper, we propose a highly accurate stereo-vision approach to complement LiDAR in autonomous robots. The system employs advanced stereo vision-based object detection to detect both tangible and non-tangible objects and then uses simple machine learning to precisely estimate the depth and size of the object. The depth and size information is then integrated into the SLAM process to enhance the robot's navigation capabilities in complex environments. Our evaluation, conducted on an autonomous robot equipped with LiDAR and stereo-vision systems demonstrates high accuracy in the estimation of an object's depth and size. A video illustration of the proposed scheme is available at: \url{https://www.youtube.com/watch?v=nusI6tA9eSk}., Comment: Accepted version of the published article in IEEE Sensors Letters