The development of large language models leads to the formation of a pre-train-then-align paradigm, in which the model is typically pre-trained on a large text corpus and undergoes a tuning stage to align the model with human preference or downstream tasks. In this work, we investigate the relationship between pre-training and fine-tuning by fine-tuning multiple intermediate pre-trained model checkpoints. Our results on 18 datasets suggest that i) continual pre-training improves the model in a latent way that unveils after fine-tuning; ii) with extra fine-tuning, the datasets that the model does not demonstrate capability gain much more than those that the model performs well during the pre-training stage; iii) although model benefits significantly through supervised fine-tuning, it may forget previously known domain knowledge and the tasks that are not seen during fine-tuning; iv) the model resembles high sensitivity to evaluation prompts after supervised fine-tuning, but this sensitivity can be alleviated by more pre-training.
Satellite images have become increasingly valuable for modelling regional climate change effects. Earth surface forecasting represents one such task that integrates satellite images with meteorological data to capture the joint evolution of regional climate change effects. However, understanding the complex relationship between specific meteorological variables and land surface evolution poses a significant challenge. In light of this challenge, our paper introduces a pipeline that integrates principles from both perturbation-based explainability techniques like LIME and global marginal explainability techniques like PDP, besides addressing the constraints of using such techniques when applying them to high-dimensional spatiotemporal deep models. The proposed pipeline simplifies the undertaking of diverse investigative analyses, such as marginal sensitivity analysis, marginal correlation analysis, lag analysis, etc., on complex land surface forecasting models In this study we utilised Convolutional Long Short-Term Memory (ConvLSTM) as the surface forecasting model and did analyses on the Normalized Difference Vegetation Index (NDVI) of the surface forecasts, since meteorological variables like temperature, pressure, and precipitation significantly influence it. The study area encompasses various regions in Europe. Our analyses show that precipitation exhibits the highest sensitivity in the study area, followed by temperature and pressure. Pressure has little to no direct effect on NDVI. Additionally, interesting nonlinear correlations between meteorological variables and NDVI have been uncovered.
Deep neural network models have a complex architecture and are overparameterized. The number of parameters is more than the whole dataset, which is highly resource-consuming. This complicates their application and limits its usage on different devices. Reduction in the number of network parameters helps to reduce the size of the model, but at the same time, thoughtlessly applied, can lead to a deterioration in the quality of the network. One way to reduce the number of model parameters is matrix decomposition, where a matrix is represented as a product of smaller matrices. In this paper, we propose a new way of applying the matrix decomposition with respect to the weights of convolutional layers. The essence of the method is to train not all convolutions, but only the subset of convolutions (basis convolutions), and represent the rest as linear combinations of the basis ones. Experiments on models from the ResNet family and the CIFAR-10 dataset demonstrate that basis convolutions can not only reduce the size of the model but also accelerate the forward and backward passes of the network. Another contribution of this work is that we propose a fast method for selecting a subset of network layers in which the use of matrix decomposition does not degrade the quality of the final model.
In this paper we propose an extension of the classical Sobol' estimator for the estimation of variance based sensitivity indices. The approach assumes a linear correlation model between the input variables which is used to decompose the contribution of an input variable into a correlated and an uncorrelated part. This method provides sampling matrices following the original joint probability distribution which are used directly to compute the model output without any assumptions or approximations of the model response function.
Run to run variability in parallel programs caused by floating-point non-associativity has been known to significantly affect reproducibility in iterative algorithms, due to accumulating errors. Non-reproducibility can critically affect the efficiency and effectiveness of correctness testing for stochastic programs. Recently, the sensitivity of deep learning training and inference pipelines to floating-point non-associativity has been found to sometimes be extreme. It can prevent certification for commercial applications, accurate assessment of robustness and sensitivity, and bug detection. New approaches in scientific computing applications have coupled deep learning models with high-performance computing, leading to an aggravation of debugging and testing challenges. Here we perform an investigation of the statistical properties of floating-point non-associativity within modern parallel programming models, and analyze performance and productivity impacts of replacing atomic operations with deterministic alternatives on GPUs. We examine the recently-added deterministic options in PyTorch within the context of GPU deployment for deep learning, uncovering and quantifying the impacts of input parameters triggering run to run variability and reporting on the reliability and completeness of the documentation. Finally, we evaluate the strategy of exploiting automatic determinism that could be provided by deterministic hardware, using the Groq accelerator for inference portions of the deep learning pipeline. We demonstrate the benefits that a hardware-based strategy can provide within reproducibility and correctness efforts.
In real case applications within the virtual prototyping process, it is not always possible to reduce the complexity of the physical models and to obtain numerical models which can be solved quickly. Usually, every single numerical simulation takes hours or even days. Although the progresses in numerical methods and high performance computing, in such cases, it is not possible to explore various model configurations, hence efficient surrogate models are required. Generally the available meta-model techniques show several advantages and disadvantages depending on the investigated problem. In this paper we present an automatic approach for the selection of the optimal suitable meta-model for the actual problem. Together with an automatic reduction of the variable space using advanced filter techniques an efficient approximation is enabled also for high dimensional problems. This filter techniques enable a reduction of the high dimensional variable space to a much smaller subspace where meta-model-based sensitivity analyses are carried out to assess the influence of important variables and to identify the optimal subspace with corresponding surrogate model which enables the most accurate probabilistic analysis. For this purpose we investigate variance-based and moment-free sensitivity measures in combination with advanced meta-models as moving least squares and kriging.
Stochastic multi-agent multi-armed bandits typically assume that the rewards from each arm follow a fixed distribution, regardless of which agent pulls the arm. However, in many real-world settings, rewards can depend on the sensitivity of each agent to their environment. In medical screening, disease detection rates can vary by test type; in preference matching, rewards can depend on user preferences; and in environmental sensing, observation quality can vary across sensors. Since past work does not specify how to allocate agents of heterogeneous but known sensitivity of these types in a stochastic bandit setting, we introduce a UCB-style algorithm, Min-Width, which aggregates information from diverse agents. In doing so, we address the joint challenges of (i) aggregating the rewards, which follow different distributions for each agent-arm pair, and (ii) coordinating the assignments of agents to arms. Min-Width facilitates efficient collaboration among heterogeneous agents, exploiting the known structure in the agents' reward functions to weight their rewards accordingly. We analyze the regret of Min-Width and conduct pseudo-synthetic and fully synthetic experiments to study the performance of different levels of information sharing. Our results confirm that the gains to modeling agent heterogeneity tend to be greater when the sensitivities are more varied across agents, while combining more information does not always improve performance.
Blood clotting is an important physiological process to suppress bleeding upon injury, but when it occurs inadvertently, it can cause thrombosis, which can lead to life threatening conditions. Hence, understanding the microscopic mechanistic factors for inadvertent, spontaneous blood clotting, in absence of a vessel breach, can help in predicting and adverting such conditions. Here, we present a minimal model -- reminiscent of the SIR model -- for the initiating stage of spontaneous blood clotting, the collective activation of blood platelets. This model predicts that in the presence of very small initial activation signals, macroscopic activation of the platelet population requires a sufficient degree of heterogeneity of platelet sensitivity. To propagate the activation signal and achieve collective activation of the bulk platelet population, it requires the presence of, possibly only few, hyper-sensitive platelets, but also a sufficient proportion of platelets with intermediate, yet higher-than-average sensitivity. A comparison with experimental results demonstrates a qualitative agreement for high platelet signalling activity.
Sepsis is a significant cause of early mortality, high healthcare costs, and disability-adjusted life years. Digital interventions like continuous cardiac monitoring can help detect early warning signs and facilitate effective interventions. This paper introduces i-CardiAx, a wearable sensor utilizing low-power high-sensitivity accelerometers to measure vital signs crucial for cardiovascular health: heart rate (HR), blood pressure (BP), and respiratory rate (RR). Data collected from 10 healthy subjects using the i-CardiAx chest patch were used to develop and evaluate lightweight vital sign measurement algorithms. The algorithms demonstrated high performance: RR (-0.11 $\pm$ 0.77 breaths\min), HR (0.82 $\pm$ 2.85 beats\min), and systolic BP (-0.08 $\pm$ 6.245 mmHg). These algorithms are embedded in an ARM Cortex-M33 processor with Bluetooth Low Energy (BLE) support, achieving inference times of 4.2 ms for HR and RR, and 8.5 ms for BP. Additionally, a multi-channel quantized Temporal Convolutional Neural (TCN) Network, trained on the open-source HiRID dataset, was developed to detect sepsis onset using digitally acquired vital signs from i-CardiAx. The quantized TCN, deployed on i-CardiAx, predicted sepsis with a median time of 8.2 hours and an energy per inference of 1.29 mJ. The i-CardiAx wearable boasts a sleep power of 0.152 mW and an average power consumption of 0.77 mW, enabling a 100 mAh battery to last approximately two weeks (432 hours) with continuous monitoring of HR, BP, and RR at 30 measurements per hour and running inference every 30 minutes. In conclusion, i-CardiAx offers an energy-efficient, high-sensitivity method for long-term cardiovascular monitoring, providing predictive alerts for sepsis and other life-threatening events.
The Hypothalamic-Pituitary-Adrenal (HPA) axis is a major neuroendocrine system, and its dysregulation is implicated in various diseases. This system also presents interesting mathematical challenges for modeling. We consider a nonlinear delay differential equation model and calculate pseudospectra of three different linearizations: a time-dependent Jacobian, linearization around the limit cycle, and dynamic mode decomposition (DMD) analysis of Koopman operators (global linearization). The time-dependent Jacobian provided insight into experimental phenomena, explaining why rats respond differently to perturbations during corticosterone secretion's upward versus downward slopes. We developed new mathematical techniques for the other two linearizations to calculate pseudospectra on Banach spaces and apply DMD to delay differential equations, respectively. These methods helped establish local and global limit cycle stability and study transients. Additionally, we discuss using pseudospectra to substantiate the model in experimental contexts and establish bio-variability via data-driven methods. This work is the first to utilize pseudospectra to explore the HPA axis.
Purpose: Artificial intelligence (AI) techniques have been extensively utilized for diagnosing and prognosis of several diseases in recent years. This study identifies, appraises and synthesizes published studies on the use of AI for the prognosis of COVID-19. Method: Electronic search was performed using Medline, Google Scholar, Scopus, Embase, Cochrane and ProQuest. Studies that examined machine learning or deep learning methods to determine the prognosis of COVID-19 using CT or chest X-ray images were included. Polled sensitivity, specificity area under the curve and diagnostic odds ratio were calculated. Result: A total of 36 articles were included; various prognosis-related issues, including disease severity, mechanical ventilation or admission to the intensive care unit and mortality, were investigated. Several AI models and architectures were employed, such as the Siamense model, support vector machine, Random Forest , eXtreme Gradient Boosting, and convolutional neural networks. The models achieved 71%, 88% and 67% sensitivity for mortality, severity assessment and need for ventilation, respectively. The specificity of 69%, 89% and 89% were reported for the aforementioned variables. Conclusion: Based on the included articles, machine learning and deep learning methods used for the prognosis of COVID-19 patients using radiomic features from CT or CXR images can help clinicians manage patients and allocate resources more effectively. These studies also demonstrate that combining patient demographic, clinical data, laboratory tests and radiomic features improves model performances.
Recently, deep learning-based single-frame infrared small target (SIRST) detection technology has made significant progress. However, existing infrared small target detection methods are often optimized for a fixed image resolution, a single wavelength, or a specific imaging system, limiting their breadth and flexibility in practical applications. Therefore, we propose a refined infrared small target detection scheme based on an adjustable sensitivity (AS) strategy and multi-scale fusion. Specifically, a multi-scale model fusion framework based on multi-scale direction-aware network (MSDA-Net) is constructed, which uses input images of multiple scales to train multiple models and fuses them. Multi-scale fusion helps characterize the shape, edge, and texture features of the target from different scales, making the model more accurate and reliable in locating the target. At the same time, we fully consider the characteristics of the infrared small target detection task and construct an edge enhancement difficulty mining (EEDM) loss. The EEDM loss helps alleviate the problem of category imbalance and guides the network to pay more attention to difficult target areas and edge features during training. In addition, we propose an adjustable sensitivity strategy for post-processing. This strategy significantly improves the detection rate of infrared small targets while ensuring segmentation accuracy. Extensive experimental results show that the proposed scheme achieves the best performance. Notably, this scheme won the first prize in the PRCV 2024 wide-area infrared small target detection competition.
Explainable Artificial Intelligence (XAI) is central to the debate on integrating Artificial Intelligence (AI) and Machine Learning (ML) algorithms into clinical practice. High-performing AI/ML models, such as ensemble learners and deep neural networks, often lack interpretability, hampering clinicians' trust in their predictions. To address this, XAI techniques are being developed to describe AI/ML predictions in human-understandable terms. One promising direction is the adaptation of sensitivity analysis (SA) and global sensitivity analysis (GSA), which inherently rank model inputs by their impact on predictions. Here, we introduce a novel delta-XAI method that provides local explanations of ML model predictions by extending the delta index, a GSA metric. The delta-XAI index assesses the impact of each feature's value on the predicted output for individual instances in both regression and classification problems. We formalize the delta-XAI index and provide code for its implementation. The delta-XAI method was evaluated on simulated scenarios using linear regression models, with Shapley values serving as a benchmark. Results showed that the delta-XAI index is generally consistent with Shapley values, with notable discrepancies in models with highly impactful or extreme feature values. The delta-XAI index demonstrated higher sensitivity in detecting dominant features and handling extreme feature values. Qualitatively, the delta-XAI provides intuitive explanations by leveraging probability density functions, making feature rankings clearer and more explainable for practitioners. Overall, the delta-XAI method appears promising for robustly obtaining local explanations of ML model predictions. Further investigations in real-world clinical settings will be conducted to evaluate its impact on AI-assisted clinical workflows.
The semantic similarity between documents of a text corpus can be visualized using map-like metaphors based on two-dimensional scatterplot layouts. These layouts result from a dimensionality reduction on the document-term matrix or a representation within a latent embedding, including topic models. Thereby, the resulting layout depends on the input data and hyperparameters of the dimensionality reduction and is therefore affected by changes in them. Furthermore, the resulting layout is affected by changes in the input data and hyperparameters of the dimensionality reduction. However, such changes to the layout require additional cognitive efforts from the user. In this work, we present a sensitivity study that analyzes the stability of these layouts concerning (1) changes in the text corpora, (2) changes in the hyperparameter, and (3) randomness in the initialization. Our approach has two stages: data measurement and data analysis. First, we derived layouts for the combination of three text corpora and six text embeddings and a grid-search-inspired hyperparameter selection of the dimensionality reductions. Afterward, we quantified the similarity of the layouts through ten metrics, concerning local and global structures and class separation. Second, we analyzed the resulting 42817 tabular data points in a descriptive statistical analysis. From this, we derived guidelines for informed decisions on the layout algorithm and highlight specific hyperparameter settings. We provide our implementation as a Git repository at https://github.com/hpicgs/Topic-Models-and-Dimensionality-Reduction-Sensitivity-Study and results as Zenodo archive at https://doi.org/10.5281/zenodo.12772898.
We present a data-driven machine-learning approach for modeling space-time socioeconomic dynamics. Through coarse-graining fine-scale observations, our modeling framework simplifies these complex systems to a set of tractable mechanistic relationships -- in the form of ordinary differential equations -- while preserving critical system behaviors. This approach allows for expedited 'what if' studies and sensitivity analyses, essential for informed policy-making. Our findings, from a case study of Baltimore, MD, indicate that this machine learning-augmented coarse-grained model serves as a powerful instrument for deciphering the complex interactions between social factors, geography, and exogenous stressors, offering a valuable asset for system forecasting and resilience planning.
The ubiquity of missing data has sparked considerable attention and focus on tabular data imputation methods. Diffusion models, recognized as the cutting-edge technique for data generation, demonstrate significant potential in tabular data imputation tasks. However, in pursuit of diversity, vanilla diffusion models often exhibit sensitivity to initialized noises, which hinders the models from generating stable and accurate imputation results. Additionally, the sparsity inherent in tabular data poses challenges for diffusion models in accurately modeling the data manifold, impacting the robustness of these models for data imputation. To tackle these challenges, this paper introduces an advanced diffusion model named Self-supervised imputation Diffusion Model (SimpDM for brevity), specifically tailored for tabular data imputation tasks. To mitigate sensitivity to noise, we introduce a self-supervised alignment mechanism that aims to regularize the model, ensuring consistent and stable imputation predictions. Furthermore, we introduce a carefully devised state-dependent data augmentation strategy within SimpDM, enhancing the robustness of the diffusion model when dealing with limited data. Extensive experiments demonstrate that SimpDM matches or outperforms state-of-the-art imputation methods across various scenarios.
Elementary Cellular Automata (ECA) are a well-studied computational universe that is, despite its simple configurations, capable of impressive computational variety. Harvesting this computation in a useful way has historically shown itself to be difficult, but if combined with reservoir computing (RC), this becomes much more feasible. Furthermore, RC and ECA enable energy-efficient AI, making the combination a promising concept for Edge AI. In this work, we contrast ECA to substrates of Partially-Local CA (PLCA) and Homogeneous Homogeneous Random Boolean Networks (HHRBN). They are, in comparison, the topological heterogeneous counterparts of ECA. This represents a step from ECA towards more biological-plausible substrates. We analyse these substrates by testing on an RC benchmark (5-bit memory), using Temporal Derrida plots to estimate the sensitivity and assess the defect collapse rate. We find that, counterintuitively, disordered topology does not necessarily mean disordered computation. There are countering computational "forces" of topology imperfections leading to a higher collapse rate (order) and yet, if accounted for, an increased sensitivity to the initial condition. These observations together suggest a shrinking critical range.
The present work provides an application of Global Sensitivity Analysis to supervised machine learning methods such as Random Forests. These methods act as black boxes, selecting features in high--dimensional data sets as to provide accurate classifiers in terms of prediction when new data are fed into the system. In supervised machine learning, predictors are generally ranked by importance based on their contribution to the final prediction. Global Sensitivity Analysis is primarily used in mathematical modelling to investigate the effect of the uncertainties of the input variables on the output. We apply it here as a novel way to rank the input features by their importance to the explainability of the data generating process, shedding light on how the response is determined by the dependence structure of its predictors. A simulation study shows that our proposal can be used to explore what advances can be achieved either in terms of efficiency, explanatory ability, or simply by way of confirming existing results.
Market definition holds significant importance in antitrust cases, yet achieving consensus on the correct approach remains elusive. As a result, analysts routinely entertain multiple market definitions to ensure the resilience of their conclusions. I propose a simple framework for conducting organized sensitivity analysis with respect to market definition. I model candidate market definitions as partially ordered and use a Hasse diagram, a directed acyclic graph representing a finite partial order, to summarize the sensitivity analysis. I use the Shapley value and the Shapley-Shubik power index to quantify the average marginal contribution of each firm in driving the conclusion. I illustrate the method's usefulness with an application to the Albertsons/Safeway (2015) merger.
Diagnosing and treating abnormalities in the wrist, specifically distal radius, and ulna fractures, is a crucial concern among children, adolescents, and young adults, with a higher incidence rate during puberty. However, the scarcity of radiologists and the lack of specialized training among medical professionals pose a significant risk to patient care. This problem is further exacerbated by the rising number of imaging studies and limited access to specialist reporting in certain regions. This highlights the need for innovative solutions to improve the diagnosis and treatment of wrist abnormalities. Automated wrist fracture detection using object detection has shown potential, but current studies mainly use two-stage detection methods with limited evidence for single-stage effectiveness. This study employs state-of-the-art single-stage deep neural network-based detection models YOLOv5, YOLOv6, YOLOv7, and YOLOv8 to detect wrist abnormalities. Through extensive experimentation, we found that these YOLO models outperform the commonly used two-stage detection algorithm, Faster R-CNN, in fracture detection. Additionally, compound-scaled variants of each YOLO model were compared, with YOLOv8m demonstrating a highest fracture detection sensitivity of 0.92 and mean average precision (mAP) of 0.95. On the other hand, YOLOv6m achieved the highest sensitivity across all classes at 0.83. Meanwhile, YOLOv8x recorded the highest mAP of 0.77 for all classes on the GRAZPEDWRI-DX pediatric wrist dataset, highlighting the potential of single-stage models for enhancing pediatric wrist imaging.