High-performance deep spiking neural networks via at-most-two-spike exponential coding.

Spiking neural networks (SNNs) provide necessary models and algorithms for neuromorphic computing. A popular way of building high-performance deep SNNs is to convert ANNs to SNNs, taking advantage of advanced and well-trained ANNs. Here we propose an ANN to SNN conversion methodology that uses a time-based coding scheme, named At-most-two-spike Exponential Coding (AEC), and a corresponding AEC spiking neuron model for ANN-SNN conversion. AEC neurons employ quantization-compensating spikes to improve coding accuracy and capacity, with each neuron generating up to two spikes within the time window. Two exponential decay functions with tunable parameters are proposed to represent the dynamic encoding thresholds, based on which pixel intensities are encoded into spike times and spike times are decoded into pixel intensities. The hyper-parameters of AEC neurons are fine-tuned based on the loss function of SNN-decoded values and ANN-activation values. In addition, we design two regularization terms for the number of spikes, providing the possibility to achieve the best trade-off between accuracy, latency and power consumption. The experimental results show that, compared to other similar methods, the proposed scheme not only obtains deep SNNs with higher accuracy, but also has more significant advantages in terms of energy efficiency and inference latency. More details can be found at https://github.com/RPDS2020/AEC.git.

A Joint Scheduling Scheme for WiFi Access TSN.

In the context of Industry 4.0, industrial production equipment needs to communicate through the industrial internet to improve the intelligence of industrial production. This requires the current communication network to have the ability of large-scale equipment access, multiple communication protocols/heterogeneous systems interoperability, and end-to-end deterministic low-latency transmission. Time-sensitive network (TSN), as a new generation of deterministic Ethernet communication technology, is the main development direction of time-critical communication technology applied in industrial environments, and Wi-Fi technology has become the main way of wireless access for users due to its advantages of high portability and mobility. Therefore, accessing WiFi in the TSN is a major development direction of the current industrial internet. In this paper, we model the scheduling problem of TSN and WiFi converged networks and propose a scheme based on a greedy strategy distributed estimation algorithm (GE) to solve the scheduling problem. Compared with the integer linear programming (ILP) algorithm and the Tabu algorithm, the algorithm implemented in this paper outperforms the other algorithms in being able to adapt to a variety of different scenarios and in scheduling optimization efficiency, especially when the amount of traffic to be deployed is large.

Improving Graph Convolutional Network with Learnable Edge Weights and Edge-Node Co-Embedding for Graph Anomaly Detection.

The era of Industry 4.0 is gradually transforming our society into a data-driven one, which can help us uncover valuable information from accumulated data, thereby improving the level of social governance. The detection of anomalies, is crucial for maintaining societal trust and fairness, yet it poses significant challenges due to the ubiquity of anomalies and the difficulty in identifying them accurately. This paper aims to enhance the performance of the current Graph Convolutional Network (GCN)-based Graph Anomaly Detection (GAD) algorithm on datasets with extremely low proportions of anomalous labels. This goal is achieved through modifying the GCN network structure and conducting feature extraction, thus fully utilizing three types of information in the graph: node label information, node feature information, and edge information. Firstly, we theoretically demonstrate the relationship between label propagation and feature convolution, indicating that the Label Propagation Algorithm (LPA) can serve as a regularization penalty term for GCN, aiding in training and enabling learnable edge weights, providing a basis for incorporating node label information into GCN networks. Secondly, we introduce a method to aggregate node and edge features, thereby incorporating edge information into GCN networks. Finally, we design different GCN trainable weights for node features and co-embedding features. This design allows different features to be projected into different spaces, greatly enhancing model expressiveness. Experimental results on the DGraph dataset demonstrate superior AUC performance compared to baseline models, highlighting the feasibility and efficacy of the proposed approach in addressing GAD tasks in the scene with extremely low proportions of anomalous data.

Deep-learning-driven dose prediction and verification for stereotactic radiosurgical treatment of isolated brain metastases.

Purpose: While deep learning has shown promise for automated radiotherapy planning, its application to the specific scenario of stereotactic radiosurgery (SRS) for brain metastases using fixed-field intensity modulated radiation therapy (IMRT) on a linear accelerator remains limited. This work aimed to develop and verify a deep learning-guided automated planning protocol tailored for this scenario. Methods: We collected 70 SRS plans for solitary brain metastases, of which 36 cases were for training and 34 for testing. Test cases were derived from two distinct clinical institutions. The envisioned automated planning process comprised (1): clinical dose prediction facilitated by deep-learning algorithms (2); transformation of the forecasted dose into executable plans via voxel-centric dose emulation (3); validation of the envisaged plan employing a precise dosimeter in conjunction with a linear accelerator. Dose prediction paradigms were established by engineering and refining two three-dimensional UNet architectures (UNet and AttUNet). Input parameters encompassed computed tomography scans from clinical plans and demarcations of the focal point alongside organs at potential risk (OARs); the ensuing output manifested as a 3D dose matrix tailored for each case under scrutiny. Results: Dose estimations rendered by both models mirrored the manual plans and adhered to clinical stipulations. As projected by the dual models, the apex and average doses for OARs did not deviate appreciably from those delineated in the manual plan (P-value≥0.05). AttUNet showed promising results compared to the foundational UNet. Predicted doses showcased a pronounced dose gradient, with peak concentrations localized within the target vicinity. The executable plans conformed to clinical dosimetric benchmarks and aligned with their associated verification assessments (100% gamma approval rate at 3 mm/3%). Conclusion: This study demonstrates an automated planning technique for fixed-field IMRT-based SRS for brain metastases. The envisaged plans met clinical requirements, were reproducible across centers, and achievable in deliveries. This represents progress toward automated paradigms for this specific scenario.

Pyrazine as a More Efficient Luminophore than Benzene for Producing Red-Shifted and Enhanced Photoluminescence.

The development of organic photoluminescent (PL) materials with red-shifted and enhanced emissions is beneficial to promoting their applications. Luminescent materials based on aromatic heterocycles (e.g., pyrazine) usually have red-shifted and enhanced photoluminescence compared with phenyl-based luminescent materials. In this work, the photoluminescence behaviors of pyrazine and its derivatives (o-dichloro-, o-dicyano-, and dichlorodicyano-substituted) are compared with those of benzene and its derivatives. All compounds exhibit fluorescence emissions ranging from blue to yellow, and the fluorescence emissions of pyrazinyl compounds are more red-shifted than those of phenyl compounds. Except for the o-dicyano-substituted compound, pyrazinyl compounds exhibit stronger fluorescence emissions than corresponding phenyl compounds in both pure substances and ethanol solutions. In addition, both 5,6-dichloro-2,3-dicyanopyrazine (P4) and 4,5-dichloro-1,2-dicyanobenzene (B4) exhibit room temperature phosphorescence, and the maximum delayed emission wavelength is red-shifted from 575 nm of B4 to 637 nm of P4. The energy gaps between the highest occupied molecular orbital and the lowest unoccupied molecular orbital of the monomers of pyrazinyl compounds are reduced by 0.07-1.37 eV compared with the monomers of phenyl compounds, which is the fundamental reason for the red-shifted emissions of the pyrazinyl compounds. Moreover, compared to B4, the smaller molecular spacing in the P4 crystal structure facilitates interlayer electron transfer and hence the formation of more extended through-space conjugation, resulting in the red-shifted emission of P4. This work proves that pyrazine is a more efficient luminophore than benzene for constructing PL compounds with longer emission wavelengths and higher quantum yields, which are important in guiding the design and preparation of organic PL materials.

Hydrogen-Bonding-Driven Nontraditional Photoluminescence of a ß-Enamino Ester.

Nontraditional luminogens (NTLs) do not contain any conventional chromophores (large π-conjugated structures), but they do show intrinsic photoluminescence. To achieve photoluminescence from NTLs, it is necessary to increase the extent of through-space conjugation (TSC) and suppress nonradiative decay. Incorporating strong physical interactions such as hydrogen bonding is an effective strategy to achieve this. In this work, we carried out comparative studies on the photoluminescence behaviors of two ß-enamino esters with similar chemical structures, namely methyl 3-aminocrotonate (MAC) and methyl (E)-3-(1-pyrrolidinyl)-2-butenoate (MPB). MAC crystal emits blue fluorescence under UV irradiation. The critical cluster concentration of MAC in ethanol solutions was determined by studying the relationship between the photoluminescence intensity (UV-visible absorbance) and concentration. Furthermore, MAC exhibits solvatochromism, and its emission wavelength redshifts as the solvent polarity increases. On the contrary, MPB is non-emissive in both solid state and solutions. Crystal structures and theoretical calculation prove that strong inter- and intramolecular hydrogen bonds lead to the formation of large amounts of TSC of MAC molecules in aggregated states. No hydrogen bonds and thus no effective TSC can be formed between or within MPB molecules, and this is the reason for its non-emissive nature. This work provides a deeper understanding of how hydrogen bonding contributes to the luminescence of NTLs.

Inhibiting Mn Migration by Sb-Pinning Transition Metal Layers in Lithium-Rich Cathode Material for Stable High-Capacity Properties.

Owing to the interacted anion and cation redox dynamics in Li2 MnO3 , the high energy density can be obtained for lithium-rich manganese-based layered transition metal (TM) oxide [Li1.2 Ni0.2 Mn0.6 O2 , LNMO]. However, irreversible migration of Mn ions and oxygen release during highly de-lithiation can destroy its layered structure, leading to voltage and capacity decline. Herein, non-TM antimony (Sb) is pinned to the TM layer of LNMO by a facile sol-gel method. High-resolution ex and in situ characterization technologies manifest that the introduction of trace Sb inhibits the migration of Mn ions, forming a more stable structure. Sb can impressively adjust the Mn-O interaction between anions and cations, beneficial to decrease the energy level of Mn 3d and O 2p orbitals and expand their band gap according to the  theoretical calculation results. As a result, the discharge specific capacity and the energy density for SbLi1.2 [Ni0.2 Mn0.6 ]O2 (SLNMO) reaches as high as 301 mAh g-1 and 1019.6 Wh kg-1 at 0.1 C, respectively. Moreover, the voltage decay is reduced by 419.8 mV compared with LNMO. The regulative interaction between Mn 3d and isolated O 2p bands provides an accurate guidance for solving electrochemical performance deficiencies of lithium-rich manganese-based cathode oxide.

Cross-domain heterogeneous residual network for single image super-resolution.

Single image super-resolution is an ill-posed problem, whose purpose is to acquire a high-resolution image from its degraded observation. Existing deep learning-based methods are compromised on their performance and speed due to the heavy design (i.e., huge model size) of networks. In this paper, we propose a novel high-performance cross-domain heterogeneous residual network for super-resolved image reconstruction. Our network models heterogeneous residuals between different feature layers by hierarchical residual learning. In outer residual learning, dual-domain enhancement modules extract the frequency-domain information to reinforce the space-domain features of network mapping. In middle residual learning, wide-activated residual-in-residual dense blocks are constructed by concatenating the outputs from previous blocks as the inputs into all subsequent blocks for better parameter efficacy. In inner residual learning, wide-activated residual attention blocks are introduced to capture direction- and location-aware feature maps. The proposed method was evaluated on four benchmark datasets, indicating that it can construct the high-quality super-resolved images and achieve the state-of-the-art performance. Code and pre-trained models are available at https://github.com/zhangyongqin/HRN.

Synthesis and crystal structure of poly[(3-amino-1,2,4-triazole)(µ3-1H-benzimidazole-5,6-di-carboxyl-ato)cobalt(II)].

The asymmetric unit of the title coordination polymer, [Co(C9H4N2O4)(C2H4N4)] n or [Co(L 1)(L 2)] n , consists of one crystallographically independent Co2+ centre, one L 1 2- ligand and one L 2 ligand (L 1 = 1H-benzimidazole-5,6-di-carb-oxy-lic acid, L 2 = 3-amino-1,2,4-triazole). The Co2+ centre is coordinated by two carboxyl-ato-O atoms from two independent L 1 2- ligands and two nitro-gen atoms from L 2 and another L 1 ligand. Thus, the metal center adopts a four-coordinate mode, forming a tetra-hedral geometry. Inter-estingly, through the combination of two L 1 2-, two L 2 ligands and two Co2+ ions, a basic repeating unit is constructed, resulting in the formation of a one-dimensional straight chain structure. These chains are further expanded to the final three-dimensional framework via N-H⋯O hydrogen-bonding inter-actions.

Improving the Antinoise Ability of DNNs via a Bio-Inspired Noise Adaptive Activation Function Rand Softplus.

Although deep neural networks (DNNs) have led to many remarkable results in cognitive tasks, they are still far from catching up with human-level cognition in antinoise capability. New research indicates how brittle and susceptible current models are to small variations in data distribution. In this letter, we study the stochasticity-resistance character of biological neurons by simulating the input-output response process of a leaky integrate-and-fire (LIF) neuron model and proposed a novel activation function, rand softplus (RSP), to model the response process. In RSP, a scale factor η is employed to mimic the stochasticity-adaptability of biological neurons, thereby enabling the antinoise capability of a DNN to be improved by the novel activation function. We validated the performance of RSP with a 19-layer residual network (ResNet) and a 19-layer visual geometry group (VGG) on facial expression recognition data sets and compared it with other popular activation functions, such as rectified linear units (ReLU), softplus, leaky ReLU (LReLU), exponential linear unit (ELU), and noisy softplus (NSP). The experimental results show that RSP is applied to VGG-19 or ResNet-19, and the average recognition accuracy under five different noise levels exceeds the other functions on both of the two facial expression data sets; in other words, RSP outperforms the other activation functions in noise resistance. Compared with the application in ResNet-19, the application of RSP in VGG-19 can improve a network's antinoise performance to a greater extent. In addition, RSP is easier to train compared to NSP because it has only one parameter to be calculated automatically according to the input data. Therefore, this work provides the deep learning community with a novel activation function that can better deal with overfitting problems.

Robust Lane-Detection Method for Low-Speed Environments.

Vision-based lane-detection methods provide low-cost density information about roads for autonomous vehicles. In this paper, we propose a robust and efficient method to expand the application of these methods to cover low-speed environments. First, the reliable region near the vehicle is initialized and a series of rectangular detection regions are dynamically constructed along the road. Then, an improved symmetrical local threshold edge extraction is introduced to extract the edge points of the lane markings based on accurate marking width limitations. In order to meet real-time requirements, a novel Bresenham line voting space is proposed to improve the process of line segment detection. Combined with straight lines, polylines, and curves, the proposed geometric fitting method has the ability to adapt to various road shapes. Finally, different status vectors and Kalman filter transfer matrices are used to track the key points of the linear and nonlinear parts of the lane. The proposed method was tested on a public database and our autonomous platform. The experimental results show that the method is robust and efficient and can meet the real-time requirements of autonomous vehicles.