Graph Neural Networks have achieved immense success for node classification with its power to explore the topological structure in graph data across many domains including social media, Ecommerce, and FinTech. However, recent studies show that GNNs are vulnerable to attacks aimed at adversely impacting their performance, e.g., on the node classification task. Existing studies of adversarial attacks on GNN focus primarily on manipulating the connectivity between existing nodes, a task that requires greater effort on the part of the attacker in real-world applications. In contrast, it is much more expedient on the part of the attacker to inject adversarial nodes, e.g., fake profiles with forged links, into existing graphs so as to reduce the performance of the GNN in classifying existing nodes.Hence, we consider a novel form of node injection poisoning attacks on graph data. We model the key steps of a node injection attack, e.g., establishing links between the injected adversarial nodes and other nodes, choosing the label of an injected node, etc. by a Markov Decision Process. We propose a novel reinforcement learning method for Node Injection Poisoning Attacks (NIPA), to sequentially modify the labels and links of the injected nodes, without changing the connectivity between existing nodes. Specifically, we introduce a hierarchical Q-learning network to manipulate the labels of the adversarial nodes and their links with other nodes in the graph, and design an appropriate reward function to guide the reinforcement learning agent to reduce the node classification performance of GNN.The results of our experiments show that NIPA is consistently more effective than the baseline node injection attack methods for poisoning graph data used to train GNN on several benchmark data sets. We further show that the graphs poisoned by NIPA are statistically similar to the original (clean) graphs, thus enabling the attacks to evade detection.
A brain-computer interface (BCI) system using elec-5 troencephalography signals provides a convenient means of com-6 munication between the human brain and a computer. Motor 7 imagery (MI), in which motor actions are mentally rehearsed with-8 out engaging in actual physical execution, has been widely used as 9 a major BCI approach. One robust algorithm that can successfully 10 cope with the individual differences in MI-related rhythmic pat-11 terns is to create diverse ensemble classifiers using the subband 12 common spatial pattern (SBCSP) method. To aggregate outputs 13 of ensemble members, this study uses fuzzy integral with parti-14 cle swarm optimization (PSO), which can regulate subject-specific 15 parameters for the assignment of optimal confidence levels for clas-16 sifiers. The proposed system combining SBCSP, fuzzy integral, and 17 PSO exhibits robust performance for offline single-trial classifica-18 tion of MI and real-time control of a robotic arm using MI. The 19 main contribution of this paper is that it represents the first attempt 20 to utilize fuzzy fusion technique to attack the individual differ-21 ences problem of MI applications in real-world noisy environment. 22 The results of this study demonstrate the practical feasibility of 23 implementing the proposed method for real-world applications. 24 Index Terms-Brain-computer interface (BCI), electroen-25 cephalography (EEG), fuzzy integral, motor imagery (MI), particle 26 swarm optimization (PSO). 27 I. INTRODUCTION 28 B RAIN-COMPUTER interfaces (BCIs) [1] based on the 29 user's voluntary modulations of electroencephalography 30 (EEG) [2] signals provide an alternative method of communica-31 tion between humans and machines. Despite the many pivotal 32
From biomechanical point of view, strike pattern plays an important role in preventing potential injury risk in running. Traditionally, strike pattern determination was conducted by using 3D motion analysis system with cameras. However, the procedure is costly and not convenient. With the rapid development of technology, sensors have been applied in sport science field lately. Therefore, this study was designed to determine the algorithm that can identify landing strategies with a wearable sensor. Six healthy male participants were recruited to perform heel and forefoot strike strategies at 7, 10, and 13 km/h speeds. The kinematic data were collected by Vicon 3D motion analysis system and 2 inertial measurement units (IMU) attached on the dorsal side of both shoes. The data of each foot strike were gathered for pitch angle and strike index analysis. Comparing the strike index from IMU with the pitch angle from Vicon system, our results showed that both signals exhibited highly correlated changes between different strike patterns in the sagittal plane (r=0.98). Based on the findings, the IMU sensors showed potential capabilities and could be extended beyond the context of sport science to other fields, including clinical applications.
Abstract-The class imbalance problem in machine learning occurs when certain classes are underrepresented relative to the others, leading to a learning bias toward the majority classes. To cope with the skewed class distribution, many learning methods featuring minority oversampling have been proposed, which are proved to be effective. To reduce information loss during feature space projection, this study proposes a novel oversampling algorithm, named minority oversampling in kernel adaptive subspaces (MOKAS), which exploits the invariant feature extraction capability of a kernel version of the adaptive subspace self-organizing maps. The synthetic instances are generated from well-trained subspaces and then their pre-images are reconstructed in the input space. Additionally, these instances characterize nonlinear structures present in the minority class data distribution and help the learning algorithms to counterbalance the skewed class distribution in a desirable manner. Experimental results on both real and synthetic data show that the proposed MOKAS is capable of modeling complex data distribution and outperforms a set of state-of-the-art oversampling algorithms.
Data from many real-world applications can be naturally represented by multi-view networks where the different views encode different types of relationships (e.g., friendship, shared interests in music, etc.) between real-world individuals or entities. There is an urgent need for methods to obtain low-dimensional, information preserving and typically nonlinear embeddings of such multi-view networks. However, most of the work on multiview learning focuses on data that lack a network structure, and most of the work on network embeddings has focused primarily on singleview networks. Against this background, we consider the multi-view network representation learning problem, i.e., the problem of constructing lowdimensional information preserving embeddings of multi-view networks. Specifically, we investigate a novel Generative Adversarial Network (GAN) framework for Multi-View Network Embedding, namely MEGAN, aimed at preserving the information from the individual network views, while accounting for connectivity across (and hence complementarity of and correlations between) different views. The results of our experiments on two real-world multi-view data sets show that the embeddings obtained using MEGAN outperform the state-of-the-art methods on node classification, link prediction and visualization tasks.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.