We initiate the study of locally differentially private (LDP) learning with public features. We define semi-feature LDP, where some features are publicly available while the remaining ones, along with the label, require protection under local differential privacy. Under semi-feature LDP, we demonstrate that the mini-max convergence rate for non-parametric regression is significantly reduced compared to that of classical LDP. Then we propose HistOfTree, an estimator that fully leverages the information contained in both public and private features. Theoretically, HistOfTree reaches the mini-max optimal convergence rate. Empirically, HistOfTree achieves superior performance on both synthetic and real data. We also explore scenarios where users have the flexibility to select features for protection manually. In such cases, we propose an estimator and a data-driven parameter tuning strategy, leading to analogous theoretical and empirical results.

Geometric Partitioning Mode (GPM) is an effective coding tool for inter prediction that splits a block into two partitions and blends their predictions. This paper presents a new coding mode, Regression-based Geometric Partitioning Mode (RGPM), which derives a sample-based blending for bi-predictions using a reconstructed template. The RGPM can enhance flexibility in splitting and blending methods compared to GPM. Moreover, two extensions of RGPM scheme are investigated: 1) extending RGPM with template matching (TM) and merge with motion vector difference (MMVD) methods; 2) extending RGPM principle to Spatial Geometric Partitioning Mode (SGPM) for intra prediction. Experimental results show that RGPM with extensions provide 0.12%, 0.24% and 0.23% average luma BD-rate savings on top of enhanced compression model (ECM) in all intra, random access and low delay configurations, respectively. The proposed RGPM is currently adopted in ECM and its two extensions are under study in exploration experiments for future ECM developments.

Video Anomaly Detection (VAD) presents a substantial challenge in the field of computer vision. The emergence of self-supervised learning has played a crucial role in tackling this challenge, with the design of self-supervised pretext tasks proving to be an exceptionally effective method. However, how to imbue the devised pretext task with a more comprehensive understanding of video content is still a problem worth exploring. In this paper, we contribute by introducing two object-level self-supervised tasks tailored for mining temporal and spatial information in videos, subsequently applied to VAD. The self-supervised pretext tasks we have formulated are as follows: (i) Frame Rotation Prediction and (ii) Frame Interval Prediction. These two pretext tasks Focus on handling abnormality of appearance and action respectively. Our approach follows an end-to-end methodology that does not depend on pre-trained models, skeleton data, or optical flow information. In our experimental evaluations, our model has demonstrated superior performance, outperforming established competitors on two publicly available benchmarks. In particular, we achieved a micro-AUROC of 86.9 on the ShanghaiTech dataset.

Previous studies yielded discouraging results for item-level locally differentially private linear regression with $s$-sparsity assumption, where the minimax rate for $nm$ samples is $\mathcal{O}(sd / nm\varepsilon^2)$. This can be challenging for high-dimensional data, where the dimension $d$ is extremely large. In this work, we investigate user-level locally differentially private sparse linear regression. We show that with $n$ users each contributing $m$ samples, the linear dependency of dimension $d$ can be eliminated, yielding an error upper bound of $\mathcal{O}(s/ nm\varepsilon^2)$. We propose a framework that first selects candidate variables and then conducts estimation in the narrowed low-dimensional space, which is extendable to general sparse estimation problems with tight error bounds. Experiments on both synthetic and real datasets demonstrate the superiority of the proposed methods. Both the theoretical and empirical results suggest that, with the same number of samples, locally private sparse estimation is better conducted when multiple samples per user are available.

Recent advances in deep neural networks (DNNs) for computer vision tasks have made intelligent analysis on edge devices more prevalent and practical. To better distribute computational load between edge devices and the cloud, a novel deep learning deployment strategy called Collaborative Intelligence (CI) has been proposed. In this strategy, features extracted from edge devices are first compressed and then transmitted to the cloud. However, it is unclear whether these compressed features have enough information to perform diverse downstream tasks. This paper focuses on the generalization of compressed features from one neural network among other object detection and instance segmentation task networks. We first propose a scheme to evaluate the generalization of features and further perform experiments on feature compression. Our experiments show that the extracted features contain enough information for other task networks and feature compression scheme for multi-task networks offers a 82.04% average bitrate saving compared to VVC.

The recent advancements in electroencepha- logram (EEG) signals classification largely center around the domain-specific solutions that hinder the algorithm cross-discipline adaptability. This study introduces a computer-aided broad learning EEG system (CABLES) for the classification of six distinct EEG domains under a unified sequential framework. Specifically, this paper proposes three novel modules namely, complex variational mode de- composition (CVMD), ensemble optimization-based featu- res selection (EOFS), and t-distributed stochastic neighbor embedding-based samples reduction (tSNE-SR) methods respectively for the realization of CABLES. Extensive expe- riments are carried out on seven different datasets from diverse disciplines using different variants of the neural network, extreme learning machine, and machine learning classifiers employing a 10-fold cross-validation strategy. Results compared with existing studies reveal that the highest classification accuracy of 99.1%, 97.8%, 94.3%, 91.5%, 98.9%, 95.3%, and 92% is achieved for the motor imagery dataset A, dataset B, slow cortical potentials, epilepsy, alcoholic, and schizophrenia EEG datasets res- pectively. The overall empirical analysis authenticates that the proposed CABLES framework outperforms the existing domain-specific methods in terms of classification accuracies and multirole adaptability, thus can be endorsed as an effective automated neural rehabilitation system.

Pulsed-latches are treated as competing sequential elements to flip-flops, mainly for their low-power and high-performance advantages. In a typical pulsed-latch system, an explicit or implicit pulse generator (PG) is used to generate the necessary clock pulse, contributing a significant amount of power consumption. To address it, a novel resonance-based power-efficient PG circuit called RPG is proposed. A power reduction up to 60% and a more stable performance in variable temperature and voltage environments are shown in 12nm Fin-FET simulations as compared with other PG circuits in typical multi-bit applications. Furthermore, a distribution method of integrating RPG into traditional designs is provided. The evaluation in a test core shows that it achieves up to 21% in clock power reduction, with less clock skew overhead and no device area loss.

In this article, a sparse-measurement-based fault location method for all fault types in the distribution networks with/without the distributed generators access is proposed. The pre- and during-fault voltage samplings at a few buses are needful to estimate the fault position via a compressed sensing algorithm. Besides, the principles of the sparse meter allocation are theoretically deduced, which are of directive significance for the practical implementation of the proposed method. Furthermore, the method possesses feasibility for the fault resistance of smaller than 50 Ω and metering noise of lower than 1%. The performance of the method is tested on a 12.66 kV, 69-bus distribution network model established in the real-time digital simulator. Simulation results verified that the proposed method is of a good performance for various scenarios.

The power output generated by wind farms (WFs) is integrated directly or indirectly into the grid through power electronic devices and the fault current is significantly different from that of conventional synchronous generators) and exhibits a limited amplitude, a controlled phase angle, rich harmonics, and strong nonlinearities. Therefore, the traditional ratio braking type differential protection may have low sensitivity or may even refuse to operate. In order to solve this problem, a pilot protection method based on Spearman's rank correlation coefficient is proposed in this article. When the system operates normally or an external fault occurs, a traversing current flows through the transmission line, so the current waveforms on both sides are the opposite. When an internal fault occurs, there are large differences in transient current waveforms on both sides. Spearman's rank correlation coefficient can be used to measure the degree of correlation between the waveforms on both sides to distinguish between faults inside and outside the zone. Compared with existing protection based on fault characteristics of WFs, the proposed method shows good performance when the power output of WFs is weak and the circuit breaker recloses with permanent failure. Both simulation results and field-testing data validate the proposed method.

BACKGROUND: Comparing and classifying functions of gene products are important in today's biomedical research. The semantic similarity derived from the Gene Ontology (GO) annotation has been regarded as one of the most widely used indicators for protein interaction. Among the various approaches proposed, those based on the vector space model are relatively simple, but their effectiveness is far from satisfying. RESULTS: We propose a Hierarchical Vector Space Model (HVSM) for computing semantic similarity between different genes or their products, which enhances the basic vector space model by introducing the relation between GO terms. Besides the directly annotated terms, HVSM also takes their ancestors and descendants related by "is_a" and "part_of" relations into account. Moreover, HVSM introduces the concept of a Certainty Factor to calibrate the semantic similarity based on the number of terms annotated to genes. To assess the performance of our method, we applied HVSM to Homo sapiens and Saccharomyces cerevisiae protein-protein interaction datasets. Compared with TCSS, Resnik, and other classic similarity measures, HVSM achieved significant improvement for distinguishing positive from negative protein interactions. We also tested its correlation with sequence, EC, and Pfam similarity using online tool CESSM. CONCLUSIONS: HVSM showed an improvement of up to 4% compared to TCSS, 8% compared to IntelliGO, 12% compared to basic VSM, 6% compared to Resnik, 8% compared to Lin, 11% compared to Jiang, 8% compared to Schlicker, and 11% compared to SimGIC using AUC scores. CESSM test showed HVSM was comparable to SimGIC, and superior to all other similarity measures in CESSM as well as TCSS. Supplementary information and the software are available at https://github.com/kejia1215/HVSM .

PV system integration with power grid has many challenges. One of these challenges, the disconnection of PV inverters from the grid under grid faults; these disconnections lead to a stability problem in larger scale of PV power plant. This paper is focused on the comparison of some Low voltage ride-through strategies. The two-stage PV inverter configuration and its conventional control methods were discussed, then the DC braking Chopper (BC) LVRT strategy was compared to avoided the MPPT operation during the LVRT operation (A-MPPT) strategy. The proposed strategies were verified by a simulation of a grid connected PV system using MATLAB/Simulink software.

In a microgrid, due to potential reverse power profiles between the renewable energy source (RES) and the loads, energy storage devices are employed to achieve high self-consumption of RES and to minimize power surplus flowing back into the main grid. This paper proposes a variable charging/discharging threshold method to manage the energy storage system. In addition, an adaptive intelligence technique (AIT) is put forward to raise the power management efficiency. A battery-ultra-capacitor hybrid energy storage system (HESS) with merits of high energy and power density is used to evaluate the proposed method with on-site-measured RES output data. Compared with the particle swarm optimization (PSO) algorithm based on the precise predicted data of the load and the RES, the results show that the proposed method can achieve better load smoothing and self-consumption of the RES without the requirement of precise load and RES forecasting.

Higher voltage DC distribution systems are being used or considered for aircraft, ship and EV/HEV automotive applications and are required for photovoltaic (PV) systems. Fault detection and location in these systems can be very challenging and electrical arcs can cause significant damage such as electrical fires. Therefore, detecting arc faults for DC systems is receiving considerable attention in the research community. Non-contact arc characteristics and arc detection mechanisms in DC systems are addressed in this paper through theoretical development and experimental tests.

Conditional random field methods (CRFs) have gained popularity for image labeling tasks in recent years. In this paper, we describe an alternative discriminative approach, by extending the large margin principle to incorporate spatial correlations among neighboring pixels. In particular, by explicitly enforcing the sub modular condition, graph-cuts is conveniently integrated as the inference engine to attain the optimal label assignment efficiently. Our approach allows learning a model with thousands of parameters, and is shown to be capable of readily incorporating higher-order scene context. Empirical studies on a variety of image datasets suggest that our approach performs competitively compared to the state-of-the-art scene labeling methods.