The latest Deep Learning (DL) models for detection and classification have achieved an unprecedented performance over classical machine learning algorithms. However, DL models are black-box methods hard to debug, inte...
详细信息
The latest Deep Learning (DL) models for detection and classification have achieved an unprecedented performance over classical machine learning algorithms. However, DL models are black-box methods hard to debug, interpret, and certify. DL alone cannot provide explanations that can be validated by a non technical audience such as end-users or domain experts. In contrast, symbolic AI systems that convert concepts into rules or symbols –such as knowledge graphs– are easier to explain. However, they present lower generalisation and scaling capabilities. A very important challenge is to fuse DL representations with expert knowledge. One way to address this challenge, as well as the performance-explainability trade-off is by leveraging the best of both streams without obviating domain expert knowledge. In this paper, we tackle such problem by considering the symbolic knowledge is expressed in form of a domain expert knowledge graph. We present the eXplainable Neural-symbolic learning (X-NeSyL) methodology, designed to learn both symbolic and deep representations, together with an explainability metric to assess the level of alignment of machine and human expert explanations. The ultimate objective is to fuse DL representations with expert domain knowledge during the learning process so it serves as a sound basis for explainability. In particular, X-NeSyL methodology involves the concrete use of two notions of explanation, both at inference and training time respectively: 1) EXPLANet: Expert-aligned eXplainable Part-based cLAssifier NETwork Architecture, a compositional convolutional neural network that makes use of symbolic representations, and 2) SHAP-Backprop, an explainable AI-informed training procedure that corrects and guides the DL process to align with such symbolic representations in form of knowledge graphs. We showcase X-NeSyL methodology using MonuMAI dataset for monument facade image classification, and demonstrate that with our approach, it is possible to i
This article's main contributions are twofold: 1) to demonstrate how to apply the general European Union's High-Level Expert Group's (EU HLEG) guidelines for trustworthy AI in practice for the domain of he...
详细信息
Most of the recent advances in crowd counting have evolved from hand-designed density estimation networks, where multi-scale features are leveraged to address the scale variation problem, but at the expense of demandi...
详细信息
The Internet of Vehicles (IoV) is an application of the Internet of things (IoT). It faces two main security problems: (1) the central server of the IoV may not be powerful enough to support the centralized authentica...
详细信息
Object detection is an important branch of image processing and computer vision, which has become a hot research issue in recent years. Accurate target detection in a video is the foundation of intelligent surveillanc...
详细信息
In this report, we summarize the first NTIRE challenge on light field (LF) image super-resolution (SR), which aims at super-resolving LF images under the standard bicubic degradation with a magnification factor of 4. ...
详细信息
The security and performance of a cryptographic algorithm can be reduced when implemented on an embedded system, such as a 32-bit microcontroller, due to the limitation of hardware resources, such as computational pre...
详细信息
The security and performance of a cryptographic algorithm can be reduced when implemented on an embedded system, such as a 32-bit microcontroller, due to the limitation of hardware resources, such as computational precision, processing power, and memory. This paper introduces the Integer Reversible Discrete Dual-Hahn Transform (IRDDHT), a novel integer-based transform designed for secure image encryption in resource-constrained embedded systems. The IRDDHT ensures lossless encryption by eliminating rounding errors common in floating-point transforms, maintaining image fidelity during encryption and decryption. We then propose a lightweight encryption algorithm based on IRDDHT, which introduces implicit diffusion and relies on parameter sensitivity to enhance security. When implemented on the ESP32 microcontroller, the proposed algorithm occupies less than 4 % of SRAM and 8.5 % of Flash memory, achieving a throughput of 170.67 kbps at the device’s maximum clock frequency. It is energy-efficient, capable of encrypting up to 129,025 grayscale images of size 256 × 256 pixels on a standard 3300 mAh battery. The algorithm ensures perfect decryption with no loss of image fidelity and demonstrates strong resistance to statistical attacks with near-ideal correlation values, as well as brute-force attacks with a large key space of 2 199 . These results indicate that the IRDDHT-based encryption method is an effective solution for secure image encryption in embedded systems.
In recent years, machine learning potentials (MLPs) have become indispensable tools in physics, chemistry, and materials science, driving the development of software packages for molecular dynamics (MD) simulations an...
详细信息
Accurate segmentation of brain tumors from multi-modal Magnetic Resonance (MR) images is essential in brain tumor diagnosis and treatment. However, due to the existence of domain shifts among different modalities, the...
详细信息
Automatically detecting software vulnerabilities is an important problem that has attracted much attention from the academic research community. However, existing vulnerability detectors still cannot achieve the vulne...
详细信息
暂无评论