Search

This study investigates the vulnerability of time series classification models to adversarial attacks, with a focus on how these models process local versus global information under such conditions. By leveraging the Normalized Auto Correlation Function (NACF), an exploration into the inclination of neural networks is conducted. It is demonstrated that regularization techniques, particularly those employing Fast Fourier Transform (FFT) methods and targeting frequency components of perturbations, markedly enhance the effectiveness of attacks. Meanwhile, the defense strategies, like noise introduction and Gaussian filtering, are shown to significantly lower the Attack Success Rate (ASR), with approaches based on noise introducing notably effective in countering high-frequency distortions. Furthermore, models designed to prioritize global information are revealed to possess greater resistance to adversarial manipulations. These results underline the importance of designing attack and defense mechanisms, informed by frequency domain analysis, as a means to considerably reinforce the resilience of neural network models against adversarial threats., Comment: 15 pages, 7 figures

This paper introduces the FinSen dataset that revolutionizes financial market analysis by integrating economic and financial news articles from 197 countries with stock market data. The dataset's extensive coverage spans 15 years from 2007 to 2023 with temporal information, offering a rich, global perspective with 160,000 records on financial market news. Our study leverages causally validated sentiment scores and LSTM models to enhance market forecast accuracy and reliability. Utilizing the FinSen dataset, we introduce an innovative Focal Calibration Loss, reducing Expected Calibration Error (ECE) to 3.34 percent with the DAN 3 model. This not only improves prediction accuracy but also aligns probabilistic forecasts closely with real outcomes, crucial for the financial sector where predicted probability is paramount. Our approach demonstrates the effectiveness of combining sentiment analysis with precise calibration techniques for trustworthy financial forecasting where the cost of misinterpretation can be high. Finsen Data can be found at [this github URL](https://github.com/EagleAdelaide/FinSen_Dataset.git)., Comment: Accepted by The 20th International Conference Advanced Data Mining and Applications 2024 (ADMA 2024)

We propose to realize quantum anomalous Hall effect (QAHE) in two-dimensional antiferromagnetic topological insulators. We consider antiferromagnetic MnBi$_2$Te$_4$ as a concrete example. In contrast to the even-layer A-type antiferromagnetic MnBi$_2$Te$_4$ that has zero Chern number due to the combined parity-time ($\mathcal{PT}$) symmetry, the system can host a nonzero Chern number by breaking this symmetry. We show that by controlling the antiferromagnetic spin configuration, for example, down/up/up/down, to break $\mathcal{PT}$ symmetry, tetralayer antiferromagnetic MnBi$_2$Te$_4$ can realize QAHE with Chern number $\mathcal{C}=-1$. Such spin configuration can be stablized by pinning the spin orientations on top and bottom layers. Furthermore, we reveal that the edge states are layer-selective and primarily locate at the boundaries of the bottom and top layers. In addition, via tuning the on-site orbital energy which determines the inverted band gap, we find tunable Chern number from $\mathcal{C}=-1$ to $\mathcal{C}=2$ and then to $\mathcal{C}=-1$. Our work not only proposes a scheme to realize Chern number tunable QAHE in antiferromagnets without net spin magnetization, but also provide a platform for layer-selective dissipationless transport devices.

In two-dimensional antiferromagnets, we identify the mixed Berry curvature as the geometrical origin of the nonreciprocal directional dichroism (NDD), which refers to the difference in light absorption with the propagation direction flipped. Such a Berry curvature is strongly tied to the uniaxial strain in accordance with the symmetry constraint, leading to a highly tunable NDD, whose sign and magnitude can be manipulated via the strain direction. As a concrete example, we demonstrate such a phenomenon in a lattice model of MnBi2Te4. The coupling between the mixed Berry curvature and strain also suggests the magnetic quadrupole of the Bloch wave packet as the macroscopic order parameter probed by the NDD in two dimensions, distinct from the multiferroic order P times M or the spin toroidal and quadrupole order within a unit cell in previous studies. Our work paves the way of the Berry-curvature engineering for optical nonreciprocity in two-dimensional antiferromagnets.

In the field of intelligent education, knowledge tracing (KT) has attracted increasing attention, which estimates and traces students' mastery of knowledge concepts to provide high-quality education. In KT, there are natural graph structures among questions and knowledge concepts so some studies explored the application of graph neural networks (GNNs) to improve the performance of the KT models which have not used graph structure. However, most of them ignored both the questions' difficulties and students' attempts at questions. Actually, questions with the same knowledge concepts have different difficulties, and students' different attempts also represent different knowledge mastery. In this paper, we propose a difficulty and attempts boosted graph-based KT (DAGKT), using rich information from students' records. Moreover, a novel method is designed to establish the question similarity relationship inspired by the F1 score. Extensive experiments on three real-world datasets demonstrate the effectiveness of the proposed DAGKT., Comment: 12 pages, 3figures, conference:ICONIP

Recently, the magnetic domain walls have been experimentally observed in antiferromagnetic topological insulators MnBi$_2$Te$_4$, where we find that the topological zero-line modes (ZLMs) appear along the domain walls. Here, we theoretically demonstrate that these ZLMs are layer-dependent in MnBi$_2$Te$_4$ multilayers. For domain walls with out-of-plane ferromagnetism, we find that ZLMs are equally distributed in the odd-number layers. When domain walls possess in-plane magnetization, the ZLMs can also exist in even-number layers due to in-plane mirror-symmetry breaking. Moreover, the conductive channels are mainly distributed in the outermost layers with increasing layer thickness. Our findings lay out a strategy in manipulating ZLMs and also can be utilized to distinguish the corresponding magnetic structures.