Search

In this study, we explore the implications of advancing AI technology on the safety of machine learning models, specifically in decision-making across diverse applications. Our research delves into the domain of network intrusion detection, covering rule-based and anomaly-based detection methods. There is a growing interest in anomaly detection within network intrusion detection systems, accompanied by an increase in adversarial attacks using maliciously crafted examples. However, the vulnerability of intrusion detection systems to backdoor attacks, a form of adversarial attack, is frequently overlooked in untrustworthy environments. This paper proposes a backdoor attack scenario, centering on the "AlertNet" intrusion detection model and utilizing the NSL-KDD dataset, a benchmark widely employed in NIDS research. The attack involves modifying features at the packet level, as network datasets are typically constructed from packets using statistical methods. Evaluation metrics include accuracy, attack success rate, baseline comparisons with clean and random data, and comparisons involving the proposed backdoor. Additionally, the study employs KL-divergence and OneClassSVM for distribution comparisons to demonstrate resilience against manual inspection by a human expert from outliers. In conclusion, the paper outlines applications and limitations and emphasizes the direction and importance of research on backdoor attacks in network intrusion detection systems. [ABSTRACT FROM AUTHOR]

Facial expression recognition (FER) is increasingly gaining importance in various emerging affective computing applications. In practice, achieving accurate FER is challenging due to the large amount of inter-personal variations such as expression intensity variations. In this paper, we propose a new spatio-temporal feature representation learning for FER that is robust to expression intensity variations. The proposed method utilizes representative expression-states (e.g., onset, apex and offset of expressions) which can be specified in facial sequences regardless of the expression intensity. The characteristics of facial expressions are encoded in two parts in this paper. As the first part, spatial image characteristics of the representative expression-state frames are learned via a convolutional neural network. Five objective terms are proposed to improve the expression class separability of the spatial feature representation. In the second part, temporal characteristics of the spatial feature representation in the first part are learned with a long short-term memory of the facial expression. Comprehensive experiments have been conducted on a deliberate expression dataset (MMI) and a spontaneous micro-expression dataset (CASME II). Experimental results showed that the proposed method achieved higher recognition rates in both datasets compared to the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

• ML-aided two-step homogenization framework of short fiber-reinforced plastics is proposed. • Series–parallel artificial neural networks were developed to 12 pseudograins from a fiber orientation tensor. • ANN model and Mori–Tanaka & Voigt homogenization were implemented into user material subroutine. • The accuracy of the proposed framework was validated with an error of about 3%. Decomposing the representative volume element (RVE) of short fiber-reinforced plastics (SFRPs) into several pseudograins (PGs) is essential for understanding its effective mechanical behavior. However, conventional PG decomposition methodologies are limited by their high computational costs due to iteration-based algorithms. To address this, we propose a machine learning-assisted PG decomposition procedure that utilizes a series–parallel artificial neural network (ANN) system to facilitate the time-consuming decomposition process. To validate the effectiveness of our proposal, we implemented a two-step homogenization framework of SFRP that consists of the series–parallel ANN system, Mori-Tanaka model, and Voigt model into ABAQUS user material subroutine (UMAT). The elastic modulus values predicted by the UMAT are found to be in good agreement with both DIGIMAT-MF and experimental values, while also maintaining low computational time. [ABSTRACT FROM AUTHOR]