Deep neural networks (DNNs) have achieved state-of-the-art performance across various applications. However, ensuring the reliability and trustworthiness of DNNs requires enhanced interpretability of model inputs and ...
Deep neural networks (DNNs) have achieved state-of-the-art performance across various applications. However, ensuring the reliability and trustworthiness of DNNs requires enhanced interpretability of model inputs and outputs. As an effective means of Explainable Artificial Intelligence (XAI) research, the interpretability of existing attribution algorithms varies depending on the choice of reference point, the quality of adversarial samples, or the applicability of gradient constraints in specific tasks. To thoroughly explore the attribution integration paths, in this paper, inspired by the iterative generation of high-quality samples in the diffusion model, we propose an Iterative Search Attribution (ISA) method. To enhance attribution accuracy, ISA distinguishes the importance of samples during gradient ascent and descent, while clipping the relatively unimportant features in the model. Specifically, we introduce a scale parameter during the iterative process to ensure the features in next iteration are always more significant than those in current iteration. Comprehensive experimental results show that our method has superior interpretability in image recognition tasks compared with state-of-the-art baselines. Our code is available at: https://***/LMBTough/ISA
Contact-less or Device-less Human Activity Recognition (HAR) using IEEE 802.11 Wireless Local Area Network (WLAN) has garnered significant interest due to its ubiquitous coverage, convenience, and privacy compared to ...
Contact-less or Device-less Human Activity Recognition (HAR) using IEEE 802.11 Wireless Local Area Network (WLAN) has garnered significant interest due to its ubiquitous coverage, convenience, and privacy compared to wearable and vision-based approaches. However, maintaining the accuracy of HAR in varying environments, ranges, and time periods remains a challenge. This work proposes a robust scheme using threshold segmentation, auto-correlation function (ACF), and a lightweight fully connected neural network (FCNN), which can maintain the HAR accuracy across different environments without the need to retrain the model. The proposed scheme is also evaluated across different transceivers’ ranges to understand its deployment constraints. The results demonstrate that the proposed scheme delivers consistent performance across different environments, ranges, and days, achieving an average HAR accuracy of over 97.25% without retraining. This greatly reduces the deployment complexity and enhances its practicality.
We initiate a study of a new model of property testing that is a hybrid of testing properties of distributions and testing properties of strings. Specifically, the new model refers to testing properties of distributio...
详细信息
The current system of scholarly publishing is often criticized for being slow, expensive, and not transparent. The rise of open access publishing as part of open science tenets, promoting transparency and collaboratio...
Biofilm-related biomaterial infections are notoriously challenging to treat and can lead to chronic infection and persisting *** date,a large body of research can be reviewed for coatings which potentially prevent bac...
详细信息
Biofilm-related biomaterial infections are notoriously challenging to treat and can lead to chronic infection and persisting *** date,a large body of research can be reviewed for coatings which potentially prevent bacterial infection while promoting implant *** only a very small number has been translated from bench to *** study provides an in-depth analysis of the stability,antibacterial mechanism,and biocompatibility of medical grade polycaprolactone(mPCL),coated with human serum albumin(HSA),the most abundant protein in blood plasma,and tannic acid(TA),a natural polyphenol with antibacterial *** docking studies demonstrated that HSA and TA interact mainly through hydrogen-bonding,ionic and hydrophobic interactions,leading to smooth and regular *** vitro bacteria adhesion testing showed that coated scaffolds maintained their antimicrobial properties over 3 days by significantly reducing *** colonization and biofilm ***,amplitude modulation-frequency modulation(AMFM)based viscoelasticity mapping and transmission electron microscopy(TEM)data suggested that HSA/TA-coatings cause morphological and mechanical changes on the outer cell membrane of *** leading to membrane disruption and cell death while proving non-toxic to human primary *** results support this antibiotic-free approach as an effective and biocompatible strategy to prevent biofilm-related biomaterial infections.
The European legislature has proposed the Digital Services Act (DSA) and Artificial Intelligence Act (AIA) to regulate platforms and Artificial Intelligence (AI) products. We review to what extent third-party audits a...
The European legislature has proposed the Digital Services Act (DSA) and Artificial Intelligence Act (AIA) to regulate platforms and Artificial Intelligence (AI) products. We review to what extent third-party audits are part of both laws and how is access to information on models and the data provided. By considering the value of third-party audits and third-party data access in an audit ecosystem, we identify a regulatory gap in that the AIA does not provide access to data for researchers and civil society. Our contributions to the literature include: (1) Defining an AI audit ecosystem incorporating compliance and oversight. (2) Highlighting a regulatory gap within the DSA and AIA regulatory framework, preventing the establishment of an AI audit ecosystem that has effective oversight by civil society and academia. (3) Emphasizing that third-party audits by research and civil society must be part of that ecosystem, we call for AIA amendments and delegated acts to include data and model access for certain AI products. Furthermore, we call for the DSA to provide NGOs and investigative journalists with data access to platforms by delegated acts and for adaptions and amendments of the AIA to provide third-party audits and data and model access, at least for high-risk systems. Regulations modeled after EU AI regulations should enable data access and third-party audits, fostering an AI audit ecosystem that promotes compliance and oversight mechanisms.
This paper presents a method applied for vibration reduction in an autonomous mechanism developed for robot-insect interaction. The mechanism performs continuous data collection of honeybee behaviour within a living h...
详细信息
ISBN:
(数字)9798350370942
ISBN:
(纸本)9798350370959
This paper presents a method applied for vibration reduction in an autonomous mechanism developed for robot-insect interaction. The mechanism performs continuous data collection of honeybee behaviour within a living honeybee colony. To ensure sufficient quality of the observations, the system has to be installed in proximity to the hive. However, as honeybees are particularly sensitive to vibrations, it is important to keep the vibrations produced by the mechanism below the values that can disturb the natural behaviour of the bees. Furthermore, avoiding vibrations improves the gathered data quality, decreases the maintenance frequency and prolonges the system service life. In this paper, we present the robotic observation mechanism’s development process of vibration control. The subspace identification method is applied to determine a model utilised to design a controller to actively reduce the resulting vibration while preserving the position tracking accuracy. The results of the experiments demonstrate a reduction of about 10dB in the vibration after applying the suggested controller.
Cameras that are mounted on almost every modern vehicle can be deflected due to the different reasons such as mechanical or environmental. This deviation of the camera from reference position can affect performance of...
详细信息
This paper presents the development of an object detection system based on the deep learning approach of computer vision to support the laparoscopic surgical robotic position control system. The system comprises two m...
详细信息
The server-less nature of Decentralized Federated Learning (DFL) requires allocating the aggregation role to specific participants in each federated round. Current DFL architectures ensure the trustworthiness of the a...
详细信息
暂无评论