We present a novel framework for pricing waterfall structures by simulating the uncertainty of the cashflow generated by the underlying assets in terms of value, time, and confidence levels. Our approach incorporates various probability distributions calibrated on the market price of the tranches at inception. The framework is fully implemented in PyTorch, leveraging its computational efficiency and automatic differentiation capabilities through Adjoint Algorithmic Differentiation (AAD). This enables efficient gradient computation for risk sensitivity analysis and optimization. The proposed methodology provides a flexible and scalable solution for pricing complex structured finance instruments under uncertainty
Deep prior-based approaches have demonstrated remarkable success in blind motion deblurring (BMD) recently. These methods, however, are often limited by the high non-convexity of the underlying optimization process in BMD, which leads to extreme sensitivity to the initial blur kernel. To address this issue, we propose a novel framework for BMD that leverages a deep generative model to encode the kernel prior and induce a better initialization for the blur kernel. Specifically, we pre-train a kernel generator based on a generative adversarial network (GAN) to aptly characterize the kernel's prior distribution, as well as a kernel initializer to provide a well-informed and high-quality starting point for kernel estimation. By combining these two components, we constrain the BMD solution within a compact latent kernel manifold, thus alleviating the aforementioned sensitivity for kernel initialization. Notably, the kernel generator and initializer are designed to be easily integrated with existing BMD methods in a plug-and-play manner, enhancing their overall performance. Furthermore, we extend our approach to tackle blind non-uniform motion deblurring without the need for additional priors, achieving state-of-the-art performance on challenging benchmark datasets. The source code is available at https://github.com/dch0319/GLKM-Deblur.
Large Language Models (LLMs) have shown remarkable performance across diverse tasks without domain-specific training, fueling interest in their potential for time-series forecasting. While LLMs have shown potential in zero-shot forecasting through prompting alone, recent studies suggest that LLMs lack inherent effectiveness in forecasting. Given these conflicting findings, a rigorous validation is essential for drawing reliable conclusions. In this paper, we evaluate the effectiveness of LLMs as zero-shot forecasters compared to state-of-the-art domain-specific models. Our experiments show that LLM-based zero-shot forecasters often struggle to achieve high accuracy due to their sensitivity to noise, underperforming even simple domain-specific models. We have explored solutions to reduce LLMs' sensitivity to noise in the zero-shot setting, but improving their robustness remains a significant challenge. Our findings suggest that rather than emphasizing zero-shot forecasting, a more promising direction would be to focus on fine-tuning LLMs to better process numerical sequences. Our experimental code is available at https://github.com/junwoopark92/revisiting-LLMs-zeroshot-forecaster.
Causal decomposition analysis aims to assess the effect of modifying risk factors on reducing social disparities in outcomes. Recently, this analysis has incorporated individual characteristics when modifying risk factors by utilizing optimal treatment regimes (OTRs). Since the newly defined individualized effects rely on the no omitted confounding assumption, developing sensitivity analyses to account for potential omitted confounding is essential. Moreover, OTRs and individualized effects are primarily based on binary risk factors, and no formal approach currently exists to benchmark the strength of omitted confounding using observed covariates for binary risk factors. To address this gap, we extend a simulation-based sensitivity analysis that simulates unmeasured confounders, addressing two sources of bias emerging from deriving OTRs and estimating individualized effects. Additionally, we propose a formal bounding strategy that benchmarks the strength of omitted confounding for binary risk factors. Using the High School Longitudinal Study 2009 (HSLS:09), we demonstrate this sensitivity analysis and benchmarking method.
Functional ANOVA (FANOVA) is a widely used variance-based sensitivity analysis tool. However, studies on functional-output FANOVA remain relatively scarce, especially for black-box computer experiments, which often involve complex and nonlinear functional-output relationships with unknown data distribution. Conventional approaches often rely on predefined basis functions or parametric structures that lack the flexibility to capture complex nonlinear relationships. Additionally, strong assumptions about the underlying data distributions further limit their ability to achieve a data-driven orthogonal effect decomposition. To address these challenges, this study proposes a functional-output orthogonal additive Gaussian process (FOAGP) to efficiently perform the data-driven orthogonal effect decomposition. By enforcing a conditional orthogonality constraint on the separable prior process, the proposed functional-output orthogonal additive kernel enables data-driven orthogonality without requiring prior distributional assumptions. The FOAGP framework also provides analytical formulations for local Sobol' indices and expected conditional variance sensitivity indices, enabling comprehensive sensitivity analysis by capturing both global and local effect significance. Validation through two simulation studies and a real case study on fuselage shape control confirms the model's effectiveness in orthogonal effect decomposition and variance decomposition, demonstrating its practical value in engineering applications.
Background. Chronic pain afflicts 20 % of the global population. A strictly biomedical mind-set leaves many sufferers chasing somatic cures and has fuelled the opioid crisis. The biopsychosocial model recognises pain subjective, multifactorial nature, yet uptake of psychosocial care remains low. We hypothesised that patients own pain narratives would predict their readiness to engage in psychotherapy. Methods. In a cross-sectional pilot, 24 chronic-pain patients recorded narrated pain stories on Painstory.science. Open questions probed perceived pain source, interference and influencing factors. Narratives were cleaned, embedded with a pretrained large-language model and entered into machine-learning classifiers that output ready/not ready probabilities. Results. The perception-domain model achieved 95.7 % accuracy (specificity = 0.80, sensitivity = 1.00, AUC = 0.90). The factors-influencing-pain model yielded 83.3 % accuracy (specificity = 0.60, sensitivity = 0.90, AUC = 0.75). Sentence count correlated with readiness for perception narratives (r = 0.54, p < .01) and factor narratives (r = 0.24, p < .05). Conclusion. Brief spoken pain narratives carry reliable signals of willingness to start psychosocial treatment. NLP-based screening could help clinicians match chronic-pain patients to appropriate interventions sooner, supporting a patient-centred biopsychosocial pathway.
Tuberculosis (TB) continues to pose a major public health challenge, particularly in high-burden regions such as Ethiopia, necessitating a more profound understanding of its transmission dynamics. In this study, we developed an SVEITRS compartmental model to investigate the transmission dynamics of TBs, utilizing real data from Ethiopia from 2011-2021. Model parameters were estimated via two methods: nonlinear least squares and maximum likelihood, with maximum likelihood providing more accurate and reliable results, as confirmed by a test case. The model's stability analysis indicated that there is a disease-free equilibrium in areas where the basic reproduction number ($\mathscr{R}_0$) is less than one. The results suggest that optimal conditions could lead to the elimination of TB. On the other hand, there is an endemic equilibrium in areas where $\mathscr{R}_0$ is greater than one, which means that the disease is still present. Sensitivity analysis revealed important factors affecting TB levels: higher natural death rates, vaccination rates, treatment rates, and disease-related death rates lower TB cases, whereas higher recruitment rates, contact rates, infection rates, and loss of vaccine protection increase its spread. These findings highlights to the necessity of enhancing vaccination, treatment, and recovery strategies while addressing drivers of transmission to achieve TB control in Ethiopia. This study provides useful advice for guiding TB control efforts and public health interventions in Ethiopia and similar regions.
This paper introduces a tamper-resistant framework for large language models (LLMs) in medical applications, utilizing quantum gradient descent (QGD) to detect malicious parameter modifications in real time. Integrated into a LLaMA-based model, QGD monitors weight amplitude distributions, identifying adversarial fine-tuning anomalies. Tests on the MIMIC and eICU datasets show minimal performance impact (accuracy: 89.1 to 88.3 on MIMIC) while robustly detecting tampering. PubMedQA evaluations confirm preserved biomedical question-answering capabilities. Compared to baselines like selective unlearning and cryptographic fingerprinting, QGD offers superior sensitivity to subtle weight changes. This quantum-inspired approach ensures secure, reliable medical AI, extensible to other high-stakes domains.
Generative Adversarial Networks (GAN) have shown potential in expanding limited medical imaging datasets. This study explores how different ratios of GAN-generated and real brain tumor MRI images impact the performance of a CNN in classifying healthy vs. tumorous scans. A DCGAN was used to create synthetic images which were mixed with real ones at various ratios to train a custom CNN. The CNN was then evaluated on a separate real-world test set. Our results indicate that the model maintains high sensitivity and precision in tumor classification, even when trained predominantly on synthetic data. When only a small portion of GAN data was added, such as 900 real images and 100 GAN images, the model achieved excellent performance, with test accuracy reaching 95.2%, and precision, recall, and F1-score all exceeding 95%. However, as the proportion of GAN images increased further, performance gradually declined. This study suggests that while GANs are useful for augmenting limited datasets especially when real data is scarce, too much synthetic data can introduce artifacts that affect the model's ability to generalize to real world cases.
This work introduces a novel framework for brain tumor segmentation leveraging pre-trained GANs and Unet architectures. By combining a global anomaly detection module with a refined mask generation network, the proposed model accurately identifies tumor-sensitive regions and iteratively enhances segmentation precision using adversarial loss constraints. Multi-modal MRI data and synthetic image augmentation are employed to improve robustness and address the challenge of limited annotated datasets. Experimental results on the BraTS dataset demonstrate the effectiveness of the approach, achieving high sensitivity and accuracy in both lesion-wise Dice and HD95 metrics than the baseline. This scalable method minimizes the dependency on fully annotated data, paving the way for practical real-world applications in clinical settings.
The rapid rise of e-commerce has transformed consumer behavior, prompting questions about how online adoption influences offline shopping. We examine whether consumers who adopt online shopping with a retailer become more price-sensitive in their subsequent offline purchases with that retailer. Using transaction-level data from a large Brazilian pet supplies retailer operating both online and offline, we compare ''adopters'' - customers who began shopping online after a period of offline-only purchasing - with ''non-adopters'' who remained offline-only. We estimate a discrete choice logit model with individual-level heterogeneity, using a novel algorithm to handle high-dimensional fixed effects and address price endogeneity. We then apply a staggered difference-in-differences approach to the estimated price elasticities and obtain the Average Treatment Effect on the Treated (ATT). We find that offline price sensitivity increases significantly after online adoption in three out of four product categories, particularly in items with low switching costs, such as pet hygiene. Counterfactual pricing simulations show that incorporating these behavioral spillovers into pricing strategies can increase firm profits by up to 4.1%. These results underscore the importance of recognizing cross-channel effects in consumer behavior and contribute to the literature on pricing and multichannel retailing by identifying online adoption as a key driver of offline price sensitivity.
This paper focuses on a key challenge in hybrid data-driven predictive control: the effect of measurement noise on Hankel matrices. While noise is handled in direct and indirect methods, hybrid approaches often overlook its impact during trajectory estimation. We propose a Noise-Tolerant Data-Driven Predictive Control (NTDPC) framework that integrates singular value decomposition to separate system dynamics from noise within reduced-order Hankel matrices. This enables accurate prediction with shorter data horizons and lower computational effort. A sensitivity index is introduced to support horizon selection under different noise levels. Simulation results indicate improved robustness and efficiency compared to existing hybrid methods.
Recent work has proposed a universal framework that integrates the well-established power synchronization control into vector current control. Using this controller for the parallel operation of grid-forming converters, and/or with loads that have strong voltage magnitude sensitivity, requires additional loops manipulating the voltage magnitude, e.g., $QV$ and $PV$ voltage-power droop controllers. This paper derives the input admittance of the resulting overall scheme. Sensitivity analyses based on the passivity index demonstrate the benefits of the proportional components of $QV$ and $PV$ control. A case study is presented where a grid-forming converter is operated in parallel with a generator.
In the era of large language models (LLMs), code benchmarks have become an important research area in software engineering and are widely used by practitioners. These benchmarks evaluate the performance of LLMs on specific code-related tasks, such as code understanding and generation. A critical step in constructing code benchmarks is the design of prompts. However, as existing code benchmarks typically rely on a single prompt template per task, they are prone to the issue of prompt sensitivity, where minor prompt variations could result in substantial performance variations, leading to unreliable evaluations of model capabilities. While previous studies have explored prompt sensitivity, their experimental designs and findings are limited to traditional natural language processing (NLP) tasks. In this paper, we present an empirical study to investigate prompt sensitivity in code benchmarks. We first propose a general framework that modifies prompt templates in a manner that preserves both their semantics and their structure as much as possible. Based on the framework, we conduct extensive experiments across eight code benchmark tasks on 10 representative open-source LLMs, with each task featuring 100 semantically similar prompt templates. We then analyze the evaluation results using various statistical metrics, focusing on both absolute and relative model performance. Our findings suggest that even slight prompt variations can lead to significant shifts in performance. Additionally, we observe that such variations can introduce inconsistencies in the performance rankings across different models. These insights highlight the need for considering prompt sensitivity when designing future code benchmarks, to ensure more reliable and accurate evaluation of LLM capabilities.
The hardness of learning a function that attains a target task relates to its input-sensitivity. For example, image classification tasks are input-insensitive as minor corruptions should not affect the classification results, whereas arithmetic and symbolic computation, which have been recently attracting interest, are highly input-sensitive as each input variable connects to the computation results. This study presents the first learning-based Quick Response (QR) code decoding and investigates learning functions of medium sensitivity. Our experiments reveal that Transformers can successfully decode QR codes, even beyond the theoretical error-correction limit, by learning the structure of embedded texts. They generalize from English-rich training data to other languages and even random strings. Moreover, we observe that the Transformer-based QR decoder focuses on data bits while ignoring error-correction bits, suggesting a decoding mechanism distinct from standard QR code readers.
This paper comprehensively surveys research trends in imitation learning for contact-rich robotic tasks. Contact-rich tasks, which require complex physical interactions with the environment, represent a central challenge in robotics due to their nonlinear dynamics and sensitivity to small positional deviations. The paper examines demonstration collection methodologies, including teaching methods and sensory modalities crucial for capturing subtle interaction dynamics. We then analyze imitation learning approaches, highlighting their applications to contact-rich manipulation. Recent advances in multimodal learning and foundation models have significantly enhanced performance in complex contact tasks across industrial, household, and healthcare domains. Through systematic organization of current research and identification of challenges, this survey provides a foundation for future advancements in contact-rich robotic manipulation.
This study evaluates a smartphone-based, deep-learning eye-tracking algorithm by comparing its performance against a commercial infrared-based eye tracker, the Tobii Pro Nano. The aim is to investigate the feasibility of appearance-based gaze estimation under realistic mobile usage conditions. Key sensitivity factors, including age, gender, vision correction, lighting conditions, device type, and head position, were systematically analysed. The appearance-based algorithm integrates a lightweight convolutional neural network (MobileNet-V3) with a recurrent structure (Long Short-Term Memory) to predict gaze coordinates from grayscale facial images. Gaze data were collected from 51 participants using dynamic visual stimuli, and accuracy was measured using Euclidean distance. The deep learning model produced a mean error of 17.76 mm, compared to 16.53 mm for the Tobii Pro Nano. While overall accuracy differences were small, the deep learning-based method was more sensitive to factors such as lighting, vision correction, and age, with higher failure rates observed under low-light conditions among participants using glasses and in older age groups. Device-specific and positional factors also influenced tracking performance. These results highlight the potential of appearance-based approaches for mobile eye tracking and offer a reference framework for evaluating gaze estimation systems across varied usage conditions.
Computational inverse problems for biomedical simulators suffer from limited data and relatively high parameter dimensionality. This often requires sensitivity analysis, where parameters of the model are ranked based on their influence on the specific quantities of interest. This is especially important for simulators used to build medical digital twins, as the amount of data is typically limited. For expensive models, such as blood flow models, emulation is employed to expedite the simulation time. Parameter ranking and fixing using sensitivity analysis are often heuristic, though, and vary with the specific application or simulator used. The present study provides an innovative solution to this problem by leveraging polynomial chaos expansions (PCEs) for both multioutput global sensitivity analysis and formal parameter identifiability. For the former, we use dimension reduction to efficiently quantify time-series sensitivity of a one-dimensional pulmonary hemodynamics model. We consider both Windkessel and structured tree boundary conditions. We then use PCEs to construct profile-likelihood confidence intervals to formally assess parameter identifiability, and show how changes in experimental design improve identifiability. Our work presents a novel approach to determining parameter identifiability and leverages a common emulation strategy for enabling profile-likelihood analysis in problems governed by partial differential equations.
In August 2020, the World Health Assembly approved the Global Strategy to eliminate cervical cancer, marking the first time that numerous countries committed to eliminating a form of cancer. China introduced the HPV vaccine in 2016 and has made significant advancements in both prevention and treatment strategies. However, due to the relatively late introduction of the vaccine, the burden of cervical cancer in China continues to rise. In light of this, we develop a compartmental model to assess the impact of the WHO's 90-70-90 strategy, along with adult catch-up vaccination, on the control of HPV-induced cervical cancer in China. We analyze the basic properties of the model and provide proofs of the local and global asymptotic stability of the equilibrium points. Additionally, a sensitivity analysis is performed, and we use the MCMC algorithm to fit the number of new cervical cancer cases and deaths in China from 1990 to 2021. The estimated basic reproduction number before and after the introduction of the HPV vaccine in China is 1.5026 (95% CI: 1.4051-1.6002) and 1.0726 (95% CI: 0.9384-1.2067), respectively. The sensitivity analysis reveals that screening, as a non-pharmaceutical intervention, plays a crucial role in controlling the spread of the disease. We apply the 90-70-90 strategy to predict the future number of new cervical cancer cases and deaths in China. The results indicate that prioritizing the 70-90 target combination is the most cost-effective approach and can achieve the goal of zero new cervical cancer cases by 2061. Finally, an optimal control model is developed to explore the best implementation strategies for HPV vaccination and screening under various plausible scenarios.
Populations can become spatially organised through chemotaxis autoattraction, wherein population members release their own chemoattractant. Standard models of this process usually assume phenotypic homogeneity, but recent studies have shed illumination on the inherent heterogeneity within populations: in terms of chemotactic behaviour, trait heterogeneity can range from the sensitivity to attractant gradients to the rate at which attractants are produced. We propose a framework that accounts for this heterogeneity, extending the standard Keller-Segel model to a non-local formulation in which the population is continuously structured across some phenotype state space. Focussing on autoattraction, we allow both the chemotactic sensitivity and the rate of attractant secretion to vary across the population and suppose members can switch between different phenotype states. We extend classical Turing-type linear stability analyses to determine the impact of phenotypic structuring on pattern formation, showing that the rate of switching influences both the critical condition for self-organisation and subsequent pattern dynamics. Scenarios in which the chemotactic sensitivity and attractant secretion are positively or negatively correlated are used to highlight the significance of these results.