PyTorchKR Papers

최신 논문

총 1,975편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2409.06271

A new paradigm for global sensitivity analysis

<div><p>Current theory of global sensitivity analysis, based on a nonlinear functional ANOVA decomposition of the random output, is limited in scope-for instance, the analysis is limited to the output's variance and the inputs have to be mutually independent-and leads to sensitivity indices the interpretation of which is not fully clear, especially interaction effects. Alternatively, sensitivity indices built for arbitrary user-defined importance measures have been proposed but a theory to define interactions in a systematic fashion and/or establish a decomposition of the total importance measure is still missing. It is shown that these important problems are solved all at once by adopting a new paradigm. By partitioning the inputs into those causing the change in the output and those which do not, arbitrary user-defined variability measures are identified with the outcomes of a factorial experiment at two levels, leading to all factorial effects without assuming any functional decomposition. To link various well-known sensitivity indices of the literature (Sobol indices and Shapley effects), weighted factorial effects are studied and utilized.</p></div>

arXiv:2409.06147

Multiclass Arrhythmia Classification using Smartwatch Photoplethysmography Signals Collected in Real-life Settings

Most deep learning models of multiclass arrhythmia classification are tested on fingertip photoplethysmographic (PPG) data, which has higher signal-to-noise ratios compared to smartwatch-derived PPG, and the best reported sensitivity value for premature atrial/ventricular contraction (PAC/PVC) detection is only 75%. To improve upon PAC/PVC detection sensitivity while maintaining high AF detection, we use multi-modal data which incorporates 1D PPG, accelerometers, and heart rate data as the inputs to a computationally efficient 1D bi-directional Gated Recurrent Unit (1D-Bi-GRU) model to detect three arrhythmia classes. We used motion-artifact prone smartwatch PPG data from the NIH-funded Pulsewatch clinical trial. Our multimodal model tested on 72 subjects achieved an unprecedented 83% sensitivity for PAC/PVC detection while maintaining a high accuracy of 97.31% for AF detection. These results outperformed the best state-of-the-art model by 20.81% for PAC/PVC and 2.55% for AF detection even while our model was computationally more efficient (14 times lighter and 2.7 faster).

arXiv:2409.03806

Mpox Screen Lite: AI-Driven On-Device Offline Mpox Screening for Low-Resource African Mpox Emergency Response

Background: The 2024 Mpox outbreak, particularly severe in Africa with clade 1b emergence, has highlighted critical gaps in diagnostic capabilities in resource-limited settings. This study aimed to develop and validate an artificial intelligence (AI)-driven, on-device screening tool for Mpox, designed to function offline in low-resource environments. Methods: We developed a YOLOv8n-based deep learning model trained on 2,700 images (900 each of Mpox, other skin conditions, and normal skin), including synthetic data. The model was validated on 360 images and tested on 540 images. A larger external validation was conducted using 1,500 independent images. Performance metrics included accuracy, precision, recall, F1-score, sensitivity, and specificity. Findings: The model demonstrated high accuracy (96%) in the final test set. For Mpox detection, it achieved 93% precision, 97% recall, and an F1-score of 95%. Sensitivity and specificity for Mpox detection were 97% and 96%, respectively. Performance remained consistent in the larger external validation, confirming the model's robustness and generalizability. Interpretation: This AI-driven screening tool offers a rapid, accurate, and scalable solution for Mpox detection in resource-constrained settings. Its offline functionality and high performance across diverse datasets suggest significant potential for improving Mpox surveillance and management, particularly in areas lacking traditional diagnostic infrastructure.

arXiv:2409.03628

Large-Area Conductor-Loaded PDMS Dielectric Composites for High-Sensitivity Wireless and Chipless Electromagnetic Temperature Sensors

Wireless electromagnetic sensing is a passive, non-destructive technique for measuring physical and chemical changes through permittivity or conductivity changes. However, the readout is limited by the sensitivity of the materials, often requiring complex sampling, particularly in the GHz range. We report capacitive dielectric temperature sensors based on polydimethylsiloxne (PDMS) loaded with 10 vol% of inexpensive, commercially available conductive fillers including copper powder (Cu), graphite powder (GP) and milled carbon fibre powder (CF). The sensors are tested in the range of 20{\deg}C to 110{\deg}C, with enhanced sensitivity from 20 to 60{\deg}C, and relative response of up to 85.5% at 200 MHz for CF loaded capacitors. Additionally, we demonstrate that operating frequency influences the relative sensing response by as much as 15.0% in loaded composite capacitors. Finally, we demonstrate the suitability of PDMS-CF capacitors as a sensing element in wirelessly coupled chipless resonant coils tuned to 6.78 MHz with a readout response, in the resonant frequency of the sensor. The wireless readout of the PDMS-CF chipless system exhibited an average sensitivity of 0.38 %{\deg}C-1 which is a 40x improvement over a pristine PDMS-based capacitive sensor and outperforms state of the art frequency-domain radio frequency (RF) temperature sensors including carbon-based composites at higher loadings. Exploiting the high sensitivity, we interrogate the sensor wirelessly using a low-cost and portable open-source NanoVNA demonstrating a relative response in the resonant frequency of the reader coil of 48.5%, with the response in good agreement with the instrumentation-grade vector network analyzers (VNAs), demonstrating that the sensors could be integrated into an inexpensive and portable measurement setup that does not rely on specialty equipment or highly trained operators.

arXiv:2409.01607

Data-driven topology design based on principal component analysis for 3D structural design problems

Topology optimization is a structural design methodology widely utilized to address engineering challenges. However, sensitivity-based topology optimization methods struggle to solve optimization problems characterized by strong non-linearity. Leveraging the sensitivity-free nature and high capacity of deep generative models, data-driven topology design (DDTD) methodology is considered an effective solution to this problem. Despite this, the training effectiveness of deep generative models diminishes when input size exceeds a threshold while maintaining high degrees of freedom is crucial for accurately characterizing complex structures. To resolve the conflict between the both, we propose DDTD based on principal component analysis (PCA). Its core idea is to replace the direct training of deep generative models with material distributions by using a principal component score matrix obtained from PCA computation and to obtain the generated material distributions with new features through the restoration process. We apply the proposed PCA-based DDTD to the problem of minimizing the maximum stress in 3D structural mechanics and demonstrate it can effectively address the current challenges faced by DDTD that fail to handle 3D structural design problems. Various experiments are conducted to demonstrate the effectiveness and practicability of the proposed PCA-based DDTD.

arXiv:2409.00543

How Does Diverse Interpretability of Textual Prompts Impact Medical Vision-Language Zero-Shot Tasks?

Recent advancements in medical vision-language pre-training (MedVLP) have significantly enhanced zero-shot medical vision tasks such as image classification by leveraging large-scale medical image-text pair pre-training. However, the performance of these tasks can be heavily influenced by the variability in textual prompts describing the categories, necessitating robustness in MedVLP models to diverse prompt styles. Yet, this sensitivity remains underexplored. In this work, we are the first to systematically assess the sensitivity of three widely-used MedVLP methods to a variety of prompts across 15 different diseases. To achieve this, we designed six unique prompt styles to mirror real clinical scenarios, which were subsequently ranked by interpretability. Our findings indicate that all MedVLP models evaluated show unstable performance across different prompt styles, suggesting a lack of robustness. Additionally, the models' performance varied with increasing prompt interpretability, revealing difficulties in comprehending complex medical concepts. This study underscores the need for further development in MedVLP methodologies to enhance their robustness to diverse zero-shot prompts.

arXiv:2409.02370

Do Large Language Models Possess Sensitive to Sentiment?

Large Language Models (LLMs) have recently displayed their extraordinary capabilities in language understanding. However, how to comprehensively assess the sentiment capabilities of LLMs continues to be a challenge. This paper investigates the ability of LLMs to detect and react to sentiment in text modal. As the integration of LLMs into diverse applications is on the rise, it becomes highly critical to comprehend their sensitivity to emotional tone, as it can influence the user experience and the efficacy of sentiment-driven tasks. We conduct a series of experiments to evaluate the performance of several prominent LLMs in identifying and responding appropriately to sentiments like positive, negative, and neutral emotions. The models' outputs are analyzed across various sentiment benchmarks, and their responses are compared with human evaluations. Our discoveries indicate that although LLMs show a basic sensitivity to sentiment, there are substantial variations in their accuracy and consistency, emphasizing the requirement for further enhancements in their training processes to better capture subtle emotional cues. Take an example in our findings, in some cases, the models might wrongly classify a strongly positive sentiment as neutral, or fail to recognize sarcasm or irony in the text. Such misclassifications highlight the complexity of sentiment analysis and the areas where the models need to be refined. Another aspect is that different LLMs might perform differently on the same set of data, depending on their architecture and training datasets. This variance calls for a more in-depth study of the factors that contribute to the performance differences and how they can be optimized.

arXiv:2409.01814공식 코드 · Papers with Code

Segmenting Object Affordances: Reproducibility and Sensitivity to Scale

Visual affordance segmentation identifies image regions of an object an agent can interact with. Existing methods re-use and adapt learning-based architectures for semantic segmentation to the affordance segmentation task and evaluate on small-size datasets. However, experimental setups are often not reproducible, thus leading to unfair and inconsistent comparisons. In this work, we benchmark these methods under a reproducible setup on two single objects scenarios, tabletop without occlusions and hand-held containers, to facilitate future comparisons. We include a version of a recent architecture, Mask2Former, re-trained for affordance segmentation and show that this model is the best-performing on most testing sets of both scenarios. Our analysis shows that models are not robust to scale variations when object resolutions differ from those in the training set.

arXiv:2408.17115

Multi-centric AI Model for Unruptured Intracranial Aneurysm Detection and Volumetric Segmentation in 3D TOF-MRI

Purpose: To develop an open-source nnU-Net-based AI model for combined detection and segmentation of unruptured intracranial aneurysms (UICA) in 3D TOF-MRI, and compare models trained on datasets with aneurysm-like differential diagnoses. Methods: This retrospective study (2020-2023) included 385 anonymized 3D TOF-MRI images from 364 patients (mean age 59 years, 60% female) at multiple centers plus 113 subjects from the ADAM challenge. Images featured untreated or possible UICAs and differential diagnoses. Four distinct training datasets were created, and the nnU-Net framework was used for model development. Performance was assessed on a separate test set using sensitivity and False Positive (FP)/case rate for detection, and DICE score and NSD (Normalized Surface Distance) with a 0.5mm threshold for segmentation. Statistical analysis included chi-square, Mann-Whitney-U, and Kruskal-Wallis tests, with significance set at p < 0.05. Results: Models achieved overall sensitivity between 82% and 85% and a FP/case rate of 0.20 to 0.31, with no significant differences (p = 0.90 and p = 0.16). The primary model showed 85% sensitivity and 0.23 FP/case rate, outperforming the ADAM-challenge winner (61%) and a nnU-Net trained on ADAM data (51%) in sensitivity (p < 0.05). It achieved a mean DICE score of 0.73 and an NSD of 0.84 for correctly detected UICA. Conclusions: Our open-source, nnU-Net-based AI model (available at 10.5281/zenodo.13386859) demonstrates high sensitivity, low false positive rates, and consistent segmentation accuracy for UICA detection and segmentation in 3D TOF-MRI, suggesting its potential to improve clinical diagnosis and for monitoring of UICA.

arXiv:2408.16330

Sensitivity Analysis for Dynamic Discrete Choice Models

In dynamic discrete choice models, some parameters, such as the discount factor, are being fixed instead of being estimated. This paper proposes two sensitivity analysis procedures for dynamic discrete choice models with respect to the fixed parameters. First, I develop a local sensitivity measure that estimates the change in the target parameter for a unit change in the fixed parameter. This measure is fast to compute as it does not require model re-estimation. Second, I propose a global sensitivity analysis procedure that uses model primitives to study the relationship between target parameters and fixed parameters. I show how to apply the sensitivity analysis procedures of this paper through two empirical applications.

arXiv:2408.16194

Highly Sensitive Differential Microwave Sensor Using Enhanced Spiral Resonators for Precision Permittivity Measurement

This paper presents a highly sensitive microwave sensor for dielectric sensing. One of the main disadvantages of microwave resonant-based sensors is cross-sensitivity originated by time-dependent uncontrolled environmental factors such as temperature that affect the material under test (MUT) behavior, leading to undesirable frequency shifts and, hence, lower accuracy. However, this work eliminates the unwanted errors using the differential measurement technique by comparing two transmission resonance frequencies during a unit test setup to measure the permittivity of MUT over time. The proposed structure comprises a spiral resonator with an extended horizontal microstrip line (EH-ML) coupled to a microstrip transmission line (MTL). Creating EH-ML within the structure comprises two primary contributions: enhanced sensitivity resulting from stronger fringing fields generated by increasing the effective area and improved resolution due to higher resonance frequencies caused by a lower total capacitive coupling effect. The proposed sensor is fabricated and tested using MUTs with a permittivity of less than 80 to verify the performance. In this regard, a frequency detection resolution (FDR) of 44MHz and a sensitivity of 0.85% are achieved at a maximum permittivity of 78.3. The results of theoretical analysis, simulation, and measurement are in relatively good agreement. Consequently, the proposed highly sensitive microwave sensor offers significant advantages, such as low complexity in design and fabrication. It also offers high resolution and precision in a wide range of permittivity, which can be an attractive candidate for dielectric sensing in health, chemical and agriculture applications.

arXiv:2408.16536

Are Pose Estimators Ready for the Open World? STAGE: Synthetic Data Generation Toolkit for Auditing 3D Human Pose Estimators

The estimation of 3D human poses from images has progressed tremendously over the last few years as measured on standard benchmarks. However, performance in the open world remains underexplored, as current benchmarks cannot capture its full extent. Especially in safety-critical systems, it is crucial that 3D pose estimators are audited before deployment, and their sensitivity towards single factors or attributes occurring in the operational domain is thoroughly examined. Nevertheless, we currently lack a benchmark that would enable such fine-grained analysis. We thus present STAGE, a GenAI data toolkit for auditing 3D human pose estimators. We enable a text-to-image model to control the 3D human body pose in the generated image. This allows us to create customized annotated data covering a wide range of open-world attributes. We leverage STAGE and generate a series of benchmarks to audit the sensitivity of popular pose estimators towards attributes such as gender, ethnicity, age, clothing, location, and weather. Our results show that the presence of such naturally occurring attributes can cause severe degradation in the performance of pose estimators and leads us to question if they are ready for open-world deployment.

arXiv:2408.16227

Revisiting 360 Depth Estimation with PanoGabor: A New Fusion Perspective

Depth estimation from a monocular 360 image is important to the perception of the entire 3D environment. However, the inherent distortion and large field of view (FoV) in 360 images pose great challenges for this task. To this end, existing mainstream solutions typically introduce additional perspective-based 360 representations (\textit{e.g.}, Cubemap) to achieve effective feature extraction. Nevertheless, regardless of the introduced representations, they eventually need to be unified into the equirectangular projection (ERP) format for the subsequent depth estimation, which inevitably reintroduces the troublesome distortions. In this work, we propose an oriented distortion-aware Gabor Fusion framework (PGFuse) to address the above challenges. First, we introduce Gabor filters that analyze texture in the frequency domain, thereby extending the receptive fields and enhancing depth cues. To address the reintroduced distortions, we design a linear latitude-aware distortion representation method to generate customized, distortion-aware Gabor filters (PanoGabor filters). Furthermore, we design a channel-wise and spatial-wise unidirectional fusion module (CS-UFM) that integrates the proposed PanoGabor filters to unify other representations into the ERP format, delivering effective and distortion-free features. Considering the orientation sensitivity of the Gabor transform, we introduce a spherical gradient constraint to stabilize this sensitivity. Experimental results on three popular indoor 360 benchmarks demonstrate the superiority of the proposed PGFuse to existing state-of-the-art solutions. Code can be available upon acceptance.

arXiv:2408.11865

How Susceptible are LLMs to Influence in Prompts?

Large Language Models (LLMs) are highly sensitive to prompts, including additional context provided therein. As LLMs grow in capability, understanding their prompt-sensitivity becomes increasingly crucial for ensuring reliable and robust performance, particularly since evaluating these models becomes more challenging. In this work, we investigate how current models (Llama, Mixtral, Falcon) respond when presented with additional input from another model, mimicking a scenario where a more capable model -- or a system with access to more external information -- provides supplementary information to the target model. Across a diverse spectrum of question-answering tasks, we study how an LLM's response to multiple-choice questions changes when the prompt includes a prediction and explanation from another model. Specifically, we explore the influence of the presence of an explanation, the stated authoritativeness of the source, and the stated confidence of the supplementary input. Our findings reveal that models are strongly influenced, and when explanations are provided they are swayed irrespective of the quality of the explanation. The models are more likely to be swayed if the input is presented as being authoritative or confident, but the effect is small in size. This study underscores the significant prompt-sensitivity of LLMs and highlights the potential risks of incorporating outputs from external sources without thorough scrutiny and further validation. As LLMs continue to advance, understanding and mitigating such sensitivities will be crucial for their reliable and trustworthy deployment.

arXiv:2408.10462

High-Sensitivity and Compact Time-domain Soil Moisture Sensor Using Dispersive Phase Shifter for Complex Permittivity Measurement

This paper presents a Time-Domain Transmissometry Soil Moisture Sensor (TDT-SMS) using a Dispersive Phase Shifter (DPS), consisting of an interdigital capacitor that is loaded with a stacked 4-turn Complementary Spiral Resonator (S4-CSR). Soil moisture measurement technique of the proposed sensor is based on the complex permittivity sensing property of a DPS in time domain. Soil relative permittivity which varies with its moisture content is measured by burying the DPS under a soil mass and changing its phase difference while excited with a 114 MHz sine wave (single tone). DPS output phase and magnitude are compared with the reference signal and measured with a phase/loss detector. The proposed sensor exhibits accuracy better than +-1.2 percent at the highest Volumetric Water Content (VWC=30 percent) for sandy-type soil. Precise design guide is developed and simulations are performed to achieve a highly sensitive sensor. The measurement results validate the accuracy of theoretical analysis and design procedure. Owning the advantages of low profile, low power consumption, and high sensitivity makes the proposed TDT-SMS a good candidate for precision farming and IoT systems.

arXiv:2408.10284공식 코드 · Papers with Code

AdapMoE: Adaptive Sensitivity-based Expert Gating and Management for Efficient MoE Inference

Mixture-of-Experts (MoE) models are designed to enhance the efficiency of large language models (LLMs) without proportionally increasing the computational demands. However, their deployment on edge devices still faces significant challenges due to high on-demand loading overheads from managing sparsely activated experts. This paper introduces AdapMoE, an algorithm-system co-design framework for efficient MoE inference. AdapMoE features adaptive expert gating and management to reduce the on-demand loading overheads. We observe the heterogeneity of experts loading across layers and tokens, based on which we propose a sensitivity-based strategy to adjust the number of activated experts dynamically. Meanwhile, we also integrate advanced prefetching and cache management techniques to further reduce the loading latency. Through comprehensive evaluations on various platforms, we demonstrate AdapMoE consistently outperforms existing techniques, reducing the average number of activated experts by 25% and achieving a 1.35x speedup without accuracy degradation. Code is available at: https://github.com/PKU-SEC-Lab/AdapMoE.

arXiv:2408.10271

Data-Driven Fire Modeling: Learning First Arrival Times and Model Parameters with Neural Networks

Data-driven techniques are being increasingly applied to complement physics-based models in fire science. However, the lack of sufficiently large datasets continues to hinder the application of certain machine learning techniques. In this paper, we use simulated data to investigate the ability of neural networks to parameterize dynamics in fire science. In particular, we investigate neural networks that map five key parameters in fire spread to the first arrival time, and the corresponding inverse problem. By using simulated data, we are able to characterize the error, the required dataset size, and the convergence properties of these neural networks. For the inverse problem, we quantify the network's sensitivity in estimating each of the key parameters. The findings demonstrate the potential of machine learning in fire science, highlight the challenges associated with limited dataset sizes, and quantify the sensitivity of neural networks to estimate key parameters governing fire spread dynamics.

arXiv:2408.09208

Parametric Sensitivity Analysis for Models of Reaction Networks within Interacting Compartments

Models of reaction networks within interacting compartments (RNIC) are a generalization of stochastic reaction networks. It is most natural to think of the interacting compartments as "cells" that can appear, degrade, split, and even merge, with each cell containing an evolving copy of the underlying stochastic reaction network. Such models have a number of parameters, including those associated with the internal chemical model and those associated with the compartment interactions, and it is natural to want efficient computational methods for the numerical estimation of sensitivities of model statistics with respect to these parameters. Motivated by the extensive work on computational methods for parametric sensitivity analysis in the context of stochastic reaction networks over the past few decades, we provide a number of methods in the basic RNIC setting. Provided methods include the (unbiased) Girsanov transformation method (also called the Likelihood Ratio method) and a number of coupling methods for the implementation of finite differences, each motivated by methods from previous work related to stochastic reaction networks. We provide several numerical examples comparing the various methods in the new setting. We find that the relative performance of each method is in line with its analog in the "standard" stochastic reaction network setting. We have made all of the Matlab code used to implement the various methods freely available for download.

arXiv:2408.06780

A Sympathetic Nervous System Theory of Migraine

Migraine (MGR) ranks first among diseases in terms of years of lost healthy life in young adult and adult women. Currently, there is no theory of MGR. This paper presents a complete theory of migraine that explains its etiology, symptoms, pathology, and treatment. Migraine involves partially saturated (usually chronically high) sympathetic nervous system (SNS) activity, mainly due to higher sensitivity of the metabolic sensors that recruit it. MGR headache occurs when SNS activity is desensitized or excessive, resulting in hyperexcitability of baroreceptors, oxidative stress, and activation of pain pathways via TRPV1 channels and CGRP. The theory is supported by overwhelming evidence, and explains the properties of current MGR treatments.

arXiv:2408.07064

On Networks and their Applications: Stability of Gene Regulatory Networks and Gene Function Prediction using Autoencoders

We prove that nested canalizing functions are the minimum-sensitivity Boolean functions for any activity ratio and we determine the functional form of this boundary which has a nontrivial fractal structure. We further observe that the majority of the gene regulatory functions found in known biological networks (submitted to the Cell Collective database) lie on the line of minimum sensitivity which paradoxically remains largely in the unstable regime. Our results provide a quantitative basis for the argument that an evolutionary preference for nested canalizing functions in gene regulation (e.g., for higher robustness) and for elasticity of gene activity are sufficient for concentration of such systems near the "edge of chaos." The original structure of gene regulatory networks is unknown due to the undiscovered functions of some genes. Most gene function discovery approaches make use of unsupervised clustering or classification methods that discover and exploit patterns in gene expression profiles. However, existing knowledge in the field derives from multiple and diverse sources. Incorporating this know-how for novel gene function prediction can, therefore, be expected to improve such predictions. We here propose a function-specific novel gene discovery tool that uses a semi-supervised autoencoder. Our method is thus able to address the needs of a modern researcher whose expertise is typically confined to a specific functional domain. Lastly, the dynamics of unorthodox learning approaches like biologically plausible learning algorithms are investigated and found to exhibit a general form of Einstein relation.