Search

Abstract Defects-rich heterointerfaces integrated with adjustable crystalline phases and atom vacancies, as well as veiled dielectric-responsive character, are instrumental in electromagnetic dissipation. Conventional methods, however, constrain their delicate constructions. Herein, an innovative alternative is proposed: carrageenan-assistant cations-regulated (CACR) strategy, which induces a series of sulfides nanoparticles rooted in situ on the surface of carbon matrix. This unique configuration originates from strategic vacancy formation energy of sulfides and strong sulfides-carbon support interaction, benefiting the delicate construction of defects-rich heterostructures in MxSy/carbon composites (M-CAs). Impressively, these generated sulfur vacancies are firstly found to strengthen electron accumulation/consumption ability at heterointerfaces and, simultaneously, induct local asymmetry of electronic structure to evoke large dipole moment, ultimately leading to polarization coupling, i.e., defect-type interfacial polarization. Such “Janus effect” (Janus effect means versatility, as in the Greek two-headed Janus) of interfacial sulfur vacancies is intuitively confirmed by both theoretical and experimental investigations for the first time. Consequently, the sulfur vacancies-rich heterostructured Co/Ni-CAs displays broad absorption bandwidth of 6.76 GHz at only 1.8 mm, compared to sulfur vacancies-free CAs without any dielectric response. Harnessing defects-rich heterostructures, this one-pot CACR strategy may steer the design and development of advanced nanomaterials, boosting functionality across diverse application domains beyond electromagnetic response.

Abstract Limited data availability remains a significant challenge for Whisper’s low-resource speech recognition performance, falling short of practical application requirements. While previous studies have successfully reduced the recognition error rates of target language speech through fine-tuning, a comprehensive exploration and analysis of Whisper’s fine-tuning capabilities and the advantages and disadvantages of various fine-tuning strategies are still lacking. This paper aims to fill this gap by conducting comprehensive experimental exploration for Whisper’s low-resource speech recognition performance using five fine-tuning strategies with limited supervised data from seven low-resource languages. The results and analysis demonstrate that all fine-tuning strategies explored in this paper significantly enhance Whisper’s performance. However, different strategies vary in their suitability and practical effectiveness, highlighting the need for careful selection based on specific use cases and resources available.

Multi-modal multi-objective problems (MMOPs) have gained much attention during the last decade. These problems have two or more global or local Pareto optimal sets (PSs), some of which map to the same Pareto front (PF). This article presents a new affinity propagation clustering (APC) method based on the Multi-modal multi-objective differential evolution (MMODE) algorithm, called MMODE_AP, for the suit of CEC’2020 benchmark functions. First, two adaptive mutation strategies are adopted to balance exploration and exploitation and improve the diversity in the evolution process. Then, the affinity propagation clustering method is adopted to define the crowding degree in decision space (DS) and objective space (OS). Meanwhile, the non-dominated sorting scheme incorporates a particular crowding distance to truncate the population during the environmental selection process, which can obtain well-distributed solutions in both DS and OS. Moreover, the local PF membership of the solution is defined, and a predefined parameter is introduced to maintain of the local PSs and solutions around the global PS. Finally, the proposed algorithm is implemented on the suit of CEC’2020 benchmark functions for comparison with some MMODE algorithms. According to the experimental study results, the proposed MMODE_AP algorithm has about 20 better performance results on benchmark functions compared to its competitors in terms of reciprocal of Pareto sets proximity (rPSP), inverted generational distances (IGD) in the decision (IGDX) and objective (IGDF). The proposed algorithm can efficiently achieve the two goals, i.e., the convergence to the true local and global Pareto fronts along with better distributed Pareto solutions on the Pareto fronts.

Hydrogen peroxide (H2O2) is extensively used in production and life, functioning as an ecofriendly oxidant. However, the dominant method currently employed, the anthraquinone method, is not conducive to sustainable development. Recently, photocatalytic H2O2 synthesis has garnered significant attention as an environment‐friendly approach. Graphitic carbon nitride (g‐C3N4), a new type of photocatalyst, shows great potential for generating H2O2 due to its excellent stability, high photocatalytic activity, selectivity, easy adjustability, and low cost. However, the research on the mechanism of H2O2 photoproduction has not been fully understood, which limits its development and practical application. Herein, the recent progress on modified g‐C3N4 is summarized and the effect of modification methods on the photocatalytic activity and H2O2 generation process is discussed. The challenges and perspectives of g‐C3N4 toward H2O2 photoproduction are proposed.

The fluorophore source of red emissive oPD&CAT system CDs is the 5,14-dihydroquinoxalino[2,3-b] phenazine (DHQP), formed in the initial reaction stage. Subsequently, the DHQP was connected with carbon core. The excited electrons of the molecular state can transfer into the energy level of CDs, which leads to UV excitation red emissive intensity decreased.

Abstract Due to the COVID‐19 pandemic, many rapid antimicrobial agents have developed intensively. Carbon dots (CDs), a new type of carbon‐based nanomaterials, shows great potential against emerging infectious diseases and antimicrobial‐resistant infections due to their unique optical properties, excellent biocompatibility, and easy surface modification. With the definition of the CDs structure and properties, synthesis, and characteristic technology improvement, the research on the CDs as antimicrobial agents has made significant progress. However, the lack of high repeatable and exact preparation methods, and the regular antimicrobial activity make it far from practical application. In this review, we summarize the most recent progress and challenges of CDs antimicrobial. First, an overview of the characteristics and properties is given, and the advantage of CDs applied to antimicrobial is further discussed. Then, it focuses on research progress on antimicrobial mechanisms under different conditions, the critical factors affecting their antimicrobial activity, and the practical antimicrobial applications. Finally, the main challenges and future research perspectives of antimicrobial CDs are proposed.

Carbon dots (CDs), as a new zero-dimensional carbon-based nanomaterial with desirable optical properties, exhibit great potential for many application fields. However, the preparation technique of multiple emission CDs with high yield is difficult and complex. Therefore, exploring the large-scale and straightforward synthesis of multicolor CDs from a simple carbon source is necessary. In this work, the solvent-free method prepares a series of multicolor emission CDs from dihydroxybenzoic acid (DHBA). The maximum emission wavelengths are 408, 445, 553, 580, and 610 nm, respectively, covering the visible light region. The 2,4- and 2,6-CDs possess the longer emission wavelength caused by the 2,4-, and 2,6-DHBA easily undergo decarboxylation to form the larger sp2 domain graphitized structure. These CDs incorporated with g-C3N4 can significantly improve the photocatalytic water-splitting hydrogen production rate by extending the visible light absorption and enhancing the charge separation efficiency. The long-wavelength emission CDs can further enhance photocatalytic activity primarily by improving visible light absorption efficiency.

Abstract In order to realize the problems of non-intrusive load monitoring and decomposition (NILMD) from two aspects of load identification and load decomposition, based on the load characteristics of the database, this paper firstly analyzes and identifies the equipment composition of mixed electrical equipment group by using the load decision tree algorithm. Then, a 0–1 programming model for the equipment status identification is established, and the Particle Swarm Optimization (PSO) is used to solve the model for equipment state recognition, and the equipment operating state in the equipment group is identified. Finally, a simulation experiment is carried out for the partial data of Question A in the 6th “teddy cup” data mining challenge competition.

Although the attention-based speech recognition has achieved promising performances, the specific explanation of the intermediate representations remains a black box theory. In this paper, we use the method to visually show and explain continuous encoder outputs. We propose a human-intervened force alignment method to obtain labels for t-distributed stochastic neighbor embedding (t-SNE), and use them to better understand the attention mechanism and the recurrent representations. In addition, we combine t-SNE and canonical correlation analysis (CCA) to analyze the training dynamics of phones in the attention-based model. Experiments are carried on TIMIT and WSJ respectively. The aligned embeddings of the encoder outputs could form sequence manifolds of the ground truth labels. Figures of t-SNE embeddings visually show what representations the encoder shaped into and how the attention mechanism works for the speech recognition. The comparisons between different models, different layers, and different lengths of the utterance show that manifolds are clearer in the shape when outputs are from the deeper layer of the encoder, the shorter utterance, and models with better performances. We also observe that the same symbols from different utterances tend to gather at similar positions, which proves the consistency of our method. Further comparisons are taken between different epochs of the model using t-SNE and CCA. The results show that both the plosive and the nasal/flap phones converge quickly, while the long vowel phone converge slowly.

Summary: Molecular oxygen (O2) activation is a critical step in the low-temperature catalytic oxidation of carbon monoxide (CO). Photocatalysis is a facile way to activate O2 to superoxide radicals under mild conditions. Photothermal catalysis, which integrates the photocatalysis of semiconductors and the thermocatalysis of noble metals, has arisen as an appealing strategy to achieve low-temperature catalytic oxidation. Here, we report that Pd/PdO/CeO2 shows that the temperature of 90% CO conversion dropped by 30°C–50°C under irradiation. The apparent activation energy reduces from 64.6 kJ ⋅ mol−1 to 44.6 kJ ⋅ mol−1. Pd/PdO/TiO2 and Pd/PdO/Al2O3 are prepared and compared to understand the underlying mechanism. The results reveal that photocatalysis is vital in lowering the apparent activation energy besides the photothermal effect. The photo-generated electrons promote O2 activation to generate the superoxide anion. The hole can activate the adsorbed CO on the PdO. These active species are beneficial for lowering the CO oxidation temperature.

In this paper, we formulate a new system, named the uncertain sandwich impulsive control system with impulsive time windows. The presented system shows that the linear entry matrix of the system is indeterminate. We first investigate the exponential stability of the considered system by linear matrix inequalities (LMIs) and inequalities techniques, then extend the considered system to a more general one and further study the stability of the general system. Finally, numerical simulations are delivered to demonstrate the effectiveness of the theoretical results.

Abstract A method called joint connectionist temporal classification (CTC)-attention-based speech recognition has recently received increasing focus and has achieved impressive performance. A hybrid end-to-end architecture that adds an extra CTC loss to the attention-based model could force extra restrictions on alignments. To explore better the end-to-end models, we propose improvements to the feature extraction and attention mechanism. First, we introduce a joint model trained with nonnegative matrix factorization (NMF)-based high-level features. Then, we put forward a hybrid attention mechanism by incorporating multi-head attentions and calculating attention scores over multi-level outputs. Experiments on TIMIT indicate that the new method achieves state-of-the-art performance with our best model. Experiments on WSJ show that our method exhibits a word error rate (WER) that is only 0.2% worse in absolute value than the best referenced method, which is trained on a much larger dataset, and it beats all present end-to-end methods. Further experiments on LibriSpeech show that our method is also comparable to the state-of-the-art end-to-end system in WER.

Abstract The Robin boundary value problem for Laplace’s equation in the elliptic region (which is a forward problem) and its related inverse problem can be used to reconstruct Robin coefficients from measurements on a partial boundary (inverse problem). We present a numerical solution of the forward problem that uses a boundary integral equation method, and we propose a fast solver based on one that reduces the computational complexity to O(Nlog(N)) $\mathcal{O}(N\log (N))$, where N is the size of the data. We compute the solution of the inverse problem using a preconditioned Krylov subspace method where the preconditioner is based on a block matrix decomposition. The structure of the matrix is then exploited to solve the direct problem. Numerical examples are presented to illustrate the effectiveness of the proposed approach.

Abstract The raw log messages record extremely rich system, network, and application running dynamic information that is a good data source for abnormal detection. Log template extraction is an important prerequisite for log sequence anomaly detection. The problems of the existing log template extraction methods are mostly offline, and the few online methods have insufficient F1-score in multi-source log data. In view of the shortcomings of the existing methods, an online log template extraction method called LogOHC is proposed. Firstly, the raw log messages are preprocessed, and the word distributed representation (word2vec) is used to vectorize the log messages online. Then, the online hierarchical clustering algorithm is applied, and finally, log templates are generated. The experimental analysis shows that LogOHC has a higher F1-score than the existing log template extraction methods, is suitable for multi-source log data sets, and has a shorter single-step execution time, which can meet the requirements of online real-time processing.

For the security defense in the current Intelligent Transportation System (ITS), malware is often used as the security analysis data source, but only the known attack type can be detected. A general anomaly detection framework is proposed, using log data as the analysis data source. By modeling the log template sequence as a natural language sequence and using the stacked Long Short-Term Memory (LSTM) with self-attention mechanism, the framework can effectively extract the hidden pattern of the log template sequence, and well express the dependencies inside the log template sequence. The experimental results show that the overall accuracy of log sequence anomaly detection of the detection framework is better than that of existing methods and the time cost is lower.

Abstract A transfer learning-based end-to-end speech recognition approach is presented in two levels in our framework. Firstly, a feature extraction approach combining multilingual deep neural network (DNN) training with matrix factorization algorithm is introduced to extract high-level features. Secondly, the advantage of connectionist temporal classification (CTC) is transferred to the target attention-based model through a joint CTC-attention model composed of shallow recurrent neural networks (RNNs) on top of the proposed features. The experimental results show that the proposed transfer learning approach achieved the best performance among all end-to-end methods and could be comparable to the state-of-the-art speech recognition system for TIMIT when further jointly decoded with a RNN language model.

The boundedness and compactness of the inclusion mapping from Besov spaces to tent spaces are studied in this paper. Meanwhile, the boundedness, compactness, and essential norm of the Volterra integral operator Tg from Besov spaces to a class of general function spaces are also investigated.

The traditional power load identification is greatly restricted in application because of its high cost and low efficiency. In this paper, the similarity model is established to realize the noninvasive load identification of power by determining the feature database for the equipment. Firstly, the wavelet decomposition method and the wavelet threshold processing method are used to remove abnormal points and reduce noise of the original data, respectively. Secondly, the transient and steady-state characteristics of electrical equipment (active power and reactive power, harmonic current, and voltage-current trajectory) are extracted, and the feature database for the equipment is established. Thirdly, the feature similarity is defined to describe the similarity degree of any two devices under a certain feature, and the similarity model of automatic recognition of a single device is established. Finally, the device identification and calculation of power consumption are carried out for the part of data in annex 2 of question A in the 6th “teddy cup” data mining challenge competition.

The realization of a fully integrated group IV electrically driven laser at room temperature is an essential issue to be solved. We introduced a novel group IV side-emitting laser at a wavelength of 1550 nm based on a 3-layer Ge/Si quantum well (QW). By designing this scheme, we showed that the structural, electronic, and optical properties are excited for lasing at 1550 nm. The preliminary results show that the device can produce a good light spot shape convenient for direct coupling with the waveguide and single-mode light emission. The laser luminous power can reach up to 2.32 mW at a wavelength of 1550 nm with a 300-mA current. Moreover, at room temperature (300 K), the laser can maintain maximum light power and an ideal wavelength (1550 nm). Thus, this study provides a novel approach to reliable, efficient electrically pumped silicon-based lasers.

In general, the proximities to a certain diversity along the front and the Pareto front have the equal importance for solving multiobjective optimization problems (MOPs). However, most of the existing evolutionary algorithms give priority to the proximity over the diversity. To improve the diversity and decrease execution time of the nondominated sorting genetic algorithm II (NSGA-II), an improved algorithm is presented in this paper, which adopts a new vector ranking scheme to decrease the whole runtime and utilize Part and Select Algorithm (PSA) to maintain the diversity. In this algorithm, a more efficient implementation of nondominated sorting, namely, dominance degree approach for nondominated sorting (DDA-NS), is presented. Moreover, an improved diversity preservation mechanism is proposed to select a well-diversified set out of an arbitrary given set. By embedding PSA and DDA-NS into NSGA-II, denoted as DNSGA2-PSA, the whole runtime of the algorithm is decreased significantly and the exploitation of diversity is enhanced. The computational experiments show that the combination of both (DDA-NS, PSA) to NSGA-II is better than the isolated use cases, and DNSGA2-PSA still performs well in the high-dimensional cases.

Materials capable of circularly polarized luminescence (CPL) have attracted considerable attention for their promising potential applications. Bacterial cellulose (BC) was characterized as having a stable right-handed twist, which makes it a potential chiral host to endow luminophores with CPL. Then, the CPL-active BC composite film was constructed by simply impregnating bacterial cellulose pellicles with dilute aqueous solutions of luminophores (rhodamine B, carbon dots, polymer dots) and drying under ambient conditions. Simple encapsulation of luminophores renders BC with circularly polarized luminescence with a dissymmetry factor of up to 0.03. The multiple chiral centers of bacterial cellulose provide a primary asymmetric environment that can be further modulated by supramolecular chemistry, which is responsible for its circular polarization ability. We further demonstrate that commercial grade paper may endow luminophores with CPL activity, which reifies the universality of the method.

Biological pacing has been proposed as a physiologic counterpart to electronic pacing, and the sinoatrial node (SAN) is the general standard for biological pacemakers. We tested the expression of SAN pacemaker cell activity when implanted autologously in the right ventricle (RV). We induced complete heart block and implanted electronic pacemakers in the RV of adult mongrel dogs. Autologous SAN cells isolated enzymatically were studied by patch clamp to confirm SAN identity. SAN cells (400,000) were injected into the RV subepicardial free wall and dogs were monitored for 2 weeks. Pacemaker function was assessed by overdrive pacing and IV epinephrine challenge. SAN cells expressed a time-dependent inward current (I f ) activating on hyperpolarization: density = 4.3 ± 0.6 pA/pF at −105 mV. Four of the six dogs demonstrated >50% of beats originating from the implant site at 24 h. Biological pacemaker rates on days 7–14 = 45–55 bpm and post-overdrive escape times = 1.5–2.5 s. Brisk catecholamine responsiveness occurred. Dogs implanted with autologous SAN cells manifest biological pacing properties dissimilar from those of the anatomic SAN. This highlights the importance of cell and substrate interaction in generating biological pacemaker function.

The quasi-interpolation operator is widely used in numerical approximation and numerical solutions of differential equations. This paper proposes a new multiquadric(MQ) quasi-interpolate and formulates a meshfree method for the Korteweg-de Vries(KdV) equation based on the proposed multiquadric quasi-interpolate. More specifically, based on the multiquadric function, a new univariate multiquadric(MQ) quasi-interpolation scheme is structured, which possesses high accuracy, simple structure, and ease of programming. Moreover, the error estimation of the new quasi-interpolate is shown in detail. Next, a meshfree method for the Korteweg-de Vries (KdV) is proposed by using the novel multiquadric(MQ) quasi-interpolation operator. In the spatial direction, the derivative is approximated by the proposed multiquadric quasi-interpolate, and the forward divided difference approximates the temporal derivative. Several numerical examples are presented at the end of the paper to verify the expected approximation capability, and the experiment results show that the meshfree method (based on the new multiquadric(MQ) quasi-interpolation operator) is valid. [ABSTRACT FROM AUTHOR]

The extended belief rule base (EBRB) system has been widely used in decision-making problems for its accuracy and efficiency. However, EBRB system needs to traverse all the rules in the rule base and has the problems of inefficiency and inconsistency. In view of this, an extended belief rule base system inference method based on a k-dimensional (KD) tree is implemented in this paper. First, the KD tree is introduced in the construction of rule base. Then, the K-Nearest Neighbor (KNN) query optimization algorithm, based on the space indexing technique of the KD tree, is used to search for key rules. Next, the obtained key rules are activated to participate in the inference process. In addition, several experiments are conducted on function fitting, oil pipeline leakage simulation and classification datasets from UCI to verify the inference performance of the proposed method. The experimental results illustrate that the extended belief rule base system based on the KD tree can effectively improve the accuracy and stability of EBRB reasoning. [ABSTRACT FROM AUTHOR]