Search

Virtual metrology (VM) is crucial for improving process capability and production yield during semiconductor manufacturing processes. However, the performance of VM deteriorates owing to the variable operating regime and the nonlinear characteristics of the process. Herein, Variational inference Gaussian mixture model (VIGMM) and extreme learning machine (ELM) are combined to solve these issues. First, variational inference is conducted on a Gaussian mixture model to determine the number of Gaussian components automatically and the corresponding operating regimes are identified. Subsequently, an extreme learning machine is developed for each operating regime to investigate the nonlinear relationship between process inputs and outputs. Finally, VM is implemented using the corresponding local ELM, which is determined based on the responsibility of the Gaussian components. The feasibility and effectiveness of the proposed methods are validated based on a numerical case and the plasma sputtering process for fabricating thin-film transistor liquid-crystal displays. The proposed VIGMM-ELM can serve as a VM algorithm for manufacturing processes with multiple stages. [ABSTRACT FROM AUTHOR]

Self-supervised learning has shown value for uncovering informative movement features for human activity recognition. However, there has been minimal exploration of this approach for affect recognition where availability of large labelled datasets is particularly limited. In this paper, we propose a P-STEMR (Parallel Space-Time Encoding Movement Representation) architecture with the aim of addressing this gap and specifically leveraging the higher availability of human activity recognition datasets for pain-level classification. We evaluated and analyzed the architecture using three different datasets across four sets of experiments. We found statistically significant increase in average F1 score to 0.84 for pain level classification with two classes based on the architecture compared with the use of hand-crafted features. This suggests that it is capable of learning movement representations and transferring these from activity recognition based on data captured in lab settings to classification of pain levels with messier real-world data. We further found that the efficacy of transfer between datasets can be undermined by dissimilarities in population groups due to impairments that affect movement behaviour and in motion primitives (e.g. rotation versus flexion). Future work should investigate how the effect of these differences could be minimized so that data from healthy people can be more valuable for transfer learning. [ABSTRACT FROM AUTHOR]

Nowadays, metaheuristic algorithms have been emerged as a potential solution in adaptive filtering applications since they offer good convergence properties. Nonetheless, most of them fall into a local minimum since their optimization is based on a single-solution technique. As a consequence, these algorithms present a high misadjustment level and require a large population to find the optimal solution. Recently, the grey wolf optimization (GWO) algorithm has emerged as a potential solution since it requires a smaller population and possesses a stronger global optimization ability with lesser control parameters. From an engineering perspective, its compactness is an attractive feature. Therefore, this opens new horizons in the implementation of this algorithm in resource-constrained devices. In this letter, we present for the first time the use of the GWO algorithm for system identification and acoustic echo canceller (AEC) and its implementation in a field programmable gate array (FPGA) device to validate its effectiveness. Our results show that the use of the GWO algorithm achieves lower steady-state mean square error (MSE) and requires less computational resources when compared with one of the most used metaheuristic algorithm. [ABSTRACT FROM AUTHOR]

Target 3.4 of the third Sustainable Development Goal (SDG) of the United Nations (UN) General Assembly proposes to reduce premature mortality from non-communicable diseases (NCDs) by one-third. Epidemiological data presented by the World Health Organization (WHO) in 2016 show that out of a total of 57 million deaths worldwide, approximately 41 million deaths occurred due to NCDs, with 78% of such deaths occurring in low-and-middle-income countries (LMICs). The majority of investigations on NCDs agree that the leading risk factor for mortality worldwide is hypertension. Over 75% of the world's mobile phone subscriptions reside in LMICs, hence making the mobile phone particularly relevant to mHealth deployment in Africa. This study is aimed at determining the scope of the literature available on hypertension diagnosis and management in Africa, with particular emphasis on determining the feasibility, acceptability and effectiveness of interventions based on the use of mobile phones. The bulk of the evidence considered overwhelmingly shows that SMS technology is yet the most used medium for executing interventions in Africa. Consequently, the need to define novel and superior ways of providing effective and low-cost monitoring, diagnosis, and management of hypertension-related NCDs delivered through artificial intelligence and machine learning techniques is clear. [ABSTRACT FROM AUTHOR]

This article presents an anomaly detection (AD) algorithm, robust AD for rural fixed wireless LTE (RAINFOREST), to address the difficulty of fault detection in LTE networks, specifically those that are rural and fixed wireless. We propose a hybrid AD method that uses network key performance indicators (KPIs), historical KPI forecasts, density-based spatial clustering of applications with noise (DBSCAN), and statistical analysis to detect anomalies. RAINFOREST outperformed benchmark AD methods and was able to detect faults in a rural commercial fixed wireless network earlier than existing LTE threshold-based alarms. [ABSTRACT FROM AUTHOR]

The antagonistic behavior in the crowd usually exacerbates the seriousness of the situation in sudden riots, where the antagonistic emotional contagion and behavioral decision making play very important roles. However, the complex mechanism of antagonistic emotion influencing decision making, especially in the environment of sudden confrontation, has not yet been explored very clearly. In this paper, we propose an Emotional contagion-aware Deep reinforcement learning model for Antagonistic Crowd Simulation (ACSED). First, we build a group emotional contagion module based on the improved Susceptible Infected Susceptible (SIS) infection disease model, and estimate the emotional state of the group at each time step during the simulation. Then, the tendency of crowd antagonistic action is estimated based on Deep Q Network (DQN), where the agent learns the action autonomously, and leverages the mean field theory to quickly calculate the influence of other surrounding individuals on the central one. Finally, the rationality of the predicted actions by DQN is further analyzed in combination with group emotion, and the final action of the agent is determined. The proposed method in this paper is verified through several experiments with different settings. We can conclude antagonistic emotions play a critical role in the decision making of the crowd through influencing the individual behavior in the riot scenario, where individual behaviors are primarily driven by emotions and goals, rather than common rules. The experiment results also prove that the antagonistic emotion has a vital impact on the group combat, and positive emotional states are more conducive to combat. Moreover, by comparing the simulation results with real scenes, the feasibility of our method is further confirmed, which can provide good reference to formulate battle plans and improve the win rate of righteous groups in a variety of situations. [ABSTRACT FROM AUTHOR]

Recently, multi-objective evolutionary algorithms (MOEAs) have shown promising performance in terms of community detection in complex networks. However, most studies have focused on designing different strategies to achieve good community detection performance based on a single population. Unlike these studies, this study proposes a macro-micro population-based co-evolutionary multi-objective algorithm called MMCoMO for community detection in complex networks to obtain a better trade-off between exploration and exploitation. This algorithm employs two populations, i.e., macro-population and micro-population, for co-evolution to obtain better community structures. In particular, the macro-population prefers exploration and is responsible for quickly determining approximate partitions of the network to obtain good rough community structures as early as possible, whereas the micro-population favors exploitation and is responsible for searching for good fine community structures through the local search process. Thus, these two populations can be used to improve each other through interactions in the co-evolutionary process. In particular, a guiding strategy is designed using the elite (non-dominated) solutions of the macro-population to guide the micro-population, and an influencing strategy is further designed using the elite solutions of the micro-population to positively influence the macro-population. Experiments on synthetic networks and 14 real-world networks demonstrate the superiority of the proposed algorithm over several state-of-the-art community detection algorithms. [ABSTRACT FROM AUTHOR]

Cooperative co-evolution (CC) is an evolutionary algorithm that adopts the divide-and-conquer strategy to solve large-scale optimization problems. It is difficult for CC to specify a suitable subpopulation size to solve different subproblems. The population diversity may be insufficient to search for the global optimum during subpopulations' evolution. In this paper, an adaptive method for enhancing population diversity is embedded in a contribution-based CC. In CC, there are two kinds of subpopulation: the convergent or stagnant subpopulations and the non-convergent and non-stagnant subpopulations. A method is proposed in the paper to evaluate the convergent and stagnant subpopulations' contributions to improving the best overall objective value, which is different from the contribution evaluation on the non-convergent and non-stagnant subpopulations. In each co-evolutionary cycle, the new CC adaptively determines to select a subpopulation, which can make a greater contribution to improving the best overall objective value, between the above two kinds of subpopulation to undergo evolution. When a convergent or stagnant subpopulation is selected to undergo evolution, the subpopulation is re-diversified to enhance its global search capability. Our experimental results and analysis suggest that the new CC algorithm can improve the performance of CC and serves as a competitive solver for large-scale optimization problems. [ABSTRACT FROM AUTHOR]

We construct an improved and distribution-free quantile estimator (called IB quantile (IBQ) estimator) with representation as a linear combination of order statistics (OSs). The IBQ estimator can estimate quantiles of any scale-invariant population by using censored samples. Taking exponential population as an example, we derive the mean square error (mse) of the IBQ estimator under a small sample case. It is found that the IBQ estimator exhibits higher estimation efficiency when estimating large quantiles compared with the other two existing estimators. Applying the IBQ estimator to create the clutter level, we establish the IBQ constant false alarm rate (CFAR) processor. We investigate the performance of IBQ-CFAR in homogeneous background and multiple target scenarios. The theoretical analyses show that the IBQ-CFAR exhibits outstanding detection performance and false alarm control ability in multiple target scenarios. [ABSTRACT FROM AUTHOR]

Both traditional and deep-learning-based methods have limitations in extracting statistical features from polarimetric synthetic aperture radar (PolSAR) images that contain regions with different levels of heterogeneity. To address this issue, we present a statistical-texture feature learning network (STLNet) for PolSAR image classification. Our approach includes several strategies. First, we propose a novel $N$ th-order statistical feature learning ($N$ -SL) module as the statistical modeling interface to be combined with the network. In addition, we propose a multilevel high-order statistical feature learning (MSL) module based on the $N$ -SL module to represent the statistical characteristics of PolSAR images. Second, we propose a texture feature learning (TL) module to explore the spatial relationships among pixels and supplement the learned statistical features. Experimental results on the experimental synthetic aperture radar (E-SAR) and airborne synthetic aperture radar (AIRSAR) datasets demonstrate that the proposed MSL and TL modules can effectively improve classification performance. Furthermore, STLNet outperforms other networks of comparable size. [ABSTRACT FROM AUTHOR]

Finding a flat area is critical for choosing the landing site, especially in the permanently shadowed region (PSR) without reference of optical images. The digital elevation model (DEM) data are not enough to describe the complex lunar topography in the landing mission. High-resolution polarimetric synthetic aperture radar (pol-SAR) can acquire the polarized scattering map of lunar surface. In this letter, Chandrayaan-2 pol-SAR images of flat and rocky regions at non-PSR are selected with the assistance of optical image. The statistics of the Stokes parameters can tell the difference between flat and rocky surfaces. Enhanced backscattering with a large variance from rough surfaces with rocks can be distinguished from the smooth and flat areas. Large value of the first Stokes parameter $I_{1}$ with extendedly distributed probability distribution function is related with the scatterings from rough surface and rocks. A statistics rule of the Stokes parameters is established and is applied to identify the flat regions in the PSR of the crater Shoemaker. [ABSTRACT FROM AUTHOR]

Researchers analyze underground forums to study abuse and cybercrime activities. Due to the size of the forums and the domain expertise required to identify criminal discussions, most approaches employ supervised machine learning techniques to automatically classify the posts of interest. Human annotation is costly. How to select samples to annotate that account for the structure of the forum? We present a methodology to generate stratified samples based on information about the centrality properties of the population and evaluate classifier performance. We observe that by employing a sample obtained from a uniform distribution of the post degree centrality metric, we maintain the same level of precision but significantly increase the recall (+30%) compared to a sample whose distribution is respecting the population stratification. We find that classifiers trained with similar samples disagree on the classification of criminal activities up to 33% of the time when deployed on the entire forum. [ABSTRACT FROM AUTHOR]

Depression is increasingly impacting individuals both physically and psychologically worldwide. It has become a global major public health problem and attracts attention from various research fields. Traditionally, the diagnosis of depression is formulated through semi-structured interviews and supplementary questionnaires, which makes the diagnosis heavily relying on physicians’ experience and is subject to bias. However, since the pathogenic mechanism of depression is still under investigation, it is difficult for physicians to diagnose and treat, especially in the early clinical stage. As smart devices and artificial intelligence advance rapidly, understanding how depression associates with daily behaviors can be beneficial for the early stage depression diagnosis, which reduces labor costs and the likelihood of clinical mistakes as well as physicians bias. Furthermore, mental health monitoring and cloud-based remote diagnosis can be implemented through an automated depression diagnosis system. In this article, we propose an attention-based multimodality speech and text representation for depression prediction. Our model is trained to estimate the depression severity of participants using the Distress Analysis Interview Corpus-Wizard of Oz (DAIC-WOZ) dataset. For the audio modality, we use the collaborative voice analysis repository (COVAREP) features provided by the dataset and employ a Bidirectional Long Short-Term Memory Network (Bi-LSTM) followed by a Time-distributed Convolutional Neural Network (T-CNN). For the text modality, we use global vectors for word representation (GloVe) to perform word embeddings and the embeddings are fed into the Bi-LSTM network. Results show that both audio and text models perform well on the depression severity estimation task, with best sequence level $F_{1}$ F 1 score of 0.9870 and patient-level $F_{1}$ F 1 score of 0.9074 for the audio model over five classes (healthy, mild, moderate, moderately severe, and severe), as well as sequence level $F_{1}$ F 1 score of 0.9709 and patient-level $F_{1}$ F 1 score of 0.9245 for the text model over five classes. Results are similar for the multimodality fused model, with the highest $F_{1}$ F 1 score of 0.9580 on the patient-level depression detection task over five classes. Experiments show statistically significant improvements over previous works. [ABSTRACT FROM AUTHOR]

The effects of cognitive stress are complex and multi-dimensional with nuanced neural and physiological representations across our lifespan. Chronic and instantaneous stressors are known to alter both executive function and motor performance — a particularly challenging prospect for older adults. Age, sex, and motor activity are critical yet under-represented dimensions in the domain of stress detection. Through the present work, we explore a subset of these variables and the relevance of brain hemodynamics and heart rate variability (HR/V) as biomarkers of stress in an aging population. We rely on a multimodal, sex-balanced, motor-stress data set (N = 59) and an exhaustive machine learning workflow to operationalize the unique neurophysiological states that form the human stress response. We found that a quadratic discriminant was sufficient to separate the two states across feature, demographic, and activity variables. We report a stress detection accuracy between $78-98\%$ 78 - 98 % when using models trained independently on each feature-set. However, these models were highly sensitive to sex, and activity differences — with distinct regions, and features implicated in stress recognition. Both HR/V and fNIRS based features were excellent indices of cognitive stress, however neither generalized to a degree beneficial toward operational use. Our observations underscore the importance of task-context, age, and sex as factors in modeling stress detection tools for older adults. [ABSTRACT FROM AUTHOR]

There are three common beliefs about the labor market: first, that increased use of artificial intelligence (AI) is going to cause large-scale unemployment in the future; second, that the postpandemic revitalization of mobility and migration would resupply the markets affected currently by the workforce shortage; and third, that job postings, under existing employment laws, do not use biased language and offer equal opportunities to job seekers. All of them are wrong. [ABSTRACT FROM AUTHOR]

During the semiconductor manufacturing process, predicting the yield of the semiconductor is an important problem. Early detection of defective product production in the manufacturing process can save huge production cost. The data generated from the semiconductor manufacturing process have characteristics of non-normal distributions, random missing patterns and high missing rate, which complicate the prediction of the yield. We propose the Dirichlet Process - Naive Bayes model (DPNB) that can simultaneously impute missing values and address classification problems. Since the DPNB is based on the infinite Gaussian mixture model, it can estimate complex data distributions and make predictions for missing datasets with some missing patterns due to nice properties of the Gaussian distribution. The DPNB also performs well for high missing rates since it uses all information of observed components. Experiments on various real datasets including semiconductor manufacturing data show that the DPNB has better performance than state-of-the-art methods in terms of predicting missing values and target variables as percentage of missing values increases. [ABSTRACT FROM AUTHOR]

Solving constrained multi-objective optimization problems (CMOPs) is one of the most popular research topics in the multi-objective optimization community. Various approaches based on different algorithmic strategies have been proposed for benchmark CMOPs with different features and challenges. Although most of these algorithms employ one or more fixed strategies, determining the most suitable strategy is challenging and problem-dependent. This work presents a general multitasking framework comprising a main task for the original CMOP and an arbitrary number of auxiliary tasks based on different strategies for the auxiliary problems. Reinforcement learning techniques are utilized according to the population state to choose the most suitable auxiliary task during the evolutionary process. As instantiations, Q-Learning and Deep Q-Learning are successfully applied in the framework. Accordingly, two novel algorithms are developed to solve CMOPs. The results demonstrate the effectiveness of the proposed approaches on five benchmark CMOP test suites and real-world problems compared to 11 advanced methods. [ABSTRACT FROM AUTHOR]

High-speed rail station location (HSR-SL) is a complex problem with multiple conflicting factors, such as local transportation network connectivity, regional accessibility, downtown proximity, and feasibility. The proposed novel methodology quantified these factors using customized nonlinear, linear, or integer utility functions. Suitable distance decay models to quantify the local transport access to and from existing bus stops, train stations, and downtown proximity; a gravity model for potential regional accessibility to the residential population or workforce; linear models for connectivity with the existing public transit routes and normalized land cost; and a binary model with a threshold value for geographical feasibility in terms of environmental sensitivity were developed. These models were evaluated using a geospatial and network analysis-based approach, and the overall nonlinear HSR-SL problem was optimized using the particle swarm optimization algorithm. The results of a real-world study area in downtown Tokyo, Japan, revealed that customized utility functions for various factors reduced the possibility of over- or underestimation and the selection of suboptimal SLs. The proposed method improved estimation of the land cost feasibility, access to transfer points, and connectivity by 100%, 185%, and 222%, respectively, for the given case study. It was most sensitive to connectivity and proximity to the downtown area, followed by location cost, transportation access, and population or workforce potential accessibility. [ABSTRACT FROM AUTHOR]

Affect dynamics, the study of how affect develops and manifests over time, has become a popular area of research in affective computing for learning. In this article, we first provide a detailed analysis of prior affect dynamics studies, elaborating both their findings and the contextual and methodological differences between these studies. We then address methodological concerns that have not been previously addressed in the literature, discussing how various edge cases should be treated. Next, we present mathematical evidence that several past studies applied the transition metric (L) incorrectly - leading to invalid conclusions of statistical significance - and provide a corrected method. Using this corrected analysis method, we reanalyze ten past affect datasets collected in diverse contexts and synthesize the results, determining that the findings do not match the most popular theoretical model of affect dynamics. Instead, our results highlight the need to focus on cultural factors in future affect dynamics research. [ABSTRACT FROM AUTHOR]

Several optimization problems have features that hinder the capabilities of searching heuristics. To cope with this issue, different methods have been proposed to manipulate search spaces and improve the optimization process. This paper focuses on Dilation Functions (DFs), which are one of the most promising techniques to manipulate the fitness landscape, by "expanding" or "compressing" specific regions. The definition of appropriate DFs is problem dependent and requires a-priori knowledge of the optimization problem. Therefore, it is essential to introduce an automatic and efficient strategy to identify optimal DFs. With this aim, we propose a novel method based on Genetic Programming, named GP4DFs, which is capable of evolving effective DFs. GP4DFs identifies optimal dilations, where a specific DF is applied to each dimension of the search space. Moreover, thanks to a knowledge-driven initialization strategy, GP4DFs converges to better solutions with a reduced number of fitness evaluations, compared to the state-of-the-art approaches. The performance of GP4DFs is assessed on a set of 43 benchmark functions mimicking several features of real-world optimization problems. The obtained results indicate the suitability of the generated DFs. [ABSTRACT FROM AUTHOR]

Influence maximization is a crucial issue for mining the deep information of social networks, which aims to select a seed set from the network to maximize the number of influenced nodes. To evaluate the influence spread of a seed set efficiently, existing studies have proposed transformations with lower computational costs to replace the expensive Monte Carlo simulation process. These alternate transformations, based on network prior knowledge, induce different search behaviors with similar characteristics to various perspectives. Specifically, it is difficult for users to determine a suitable transformation a priori. This article proposes a multi-transformation evolutionary framework for influence maximization (MTEFIM) with convergence guarantees to exploit the potential similarities and unique advantages of alternate transformations and to avoid users manually determining the most suitable one. In MTEFIM, multiple transformations are optimized simultaneously as multiple tasks. Each transformation is assigned an evolutionary solver. Three major components of MTEFIM are conducted via: 1) estimating the potential relationship across transformations based on the degree of overlap across individuals of different populations, 2) transferring individuals across populations adaptively according to the inter-transformation relationship, and 3) selecting the final output seed set containing all the transformation's knowledge. The effectiveness of MTEFIM is validated on both benchmarks and real-world social networks. The experimental results show that MTEFIM can efficiently utilize the potentially transferable knowledge across multiple transformations to achieve highly competitive performance compared to several popular IM-specific methods. The implementation of MTEFIM can be accessed through the external link in the footnote. [ABSTRACT FROM AUTHOR]

Currently, COVID-19 is circulating in crowded places as an infectious disease. COVID-19 can be prevented from spreading rapidly in crowded areas by implementing multiple strategies. The use of unmanned aerial vehicles (UAVs) as sensing devices can be useful in detecting overcrowding events. Accordingly, in this article, we introduce a real-time system for identifying overcrowding due to events such as congestion and abnormal behavior. For the first time, a monitoring approach is proposed to detect overcrowding through the UAV and social monitoring system (SMS). We have significantly improved identification by selecting the best features from the water cycle algorithm (WCA) and making decisions based on deep transfer learning. According to the analysis of the UAV videos, the average accuracy is estimated at 96.55%. Experimental results demonstrate that the proposed approach is capable of detecting overcrowding based on UAV videos' frames and SMS's communication even in challenging conditions. [ABSTRACT FROM AUTHOR]

Objective quantification of the balancing mechanisms in humans is strongly needed in health care of older people, yet is largely missing among current clinical balance assessment methods. Hence, the main goal of this literature review is to identify methods that have the potential to meet that need. We searched in the PubMed and IEEE Xplore databases using predefined criteria, screened 1064 articles, and systematically reviewed and categorized methods from 73 studies that deal with identification of neuromuscular controller models of human upright standing from empirical data. These studies were then analyzed with the particular aim to understand to what degree such methods would be useful solutions for assessing the balance of older individuals aged above 60 years. The 16 studies that included an older subject population were especially examined with this in mind. The majority of the reviewed articles focused on research questions related to the general function of human balance control rather than clinical applicability. Further efforts need to be made to adapt these methods for more accessible and mobile technologies and to ensure that the outcomes are valid for balance assessment of a general older population. [ABSTRACT FROM AUTHOR]

This study focuses on the quasisynchronization problem for reaction–diffusion neural networks (RDNNs) in the presence of deception attacks. Under deception attacks, a time–space sampled-data (TSSD) control mechanism is proposed for RDNNs. Compared with traditional control strategies, the proposed control mechanism can not only save network bandwidth but also improve the cybersecurity of communications. Inspired by Halanay’s inequality, a new inequality is proposed, which can be effectively applied to the quasisynchronization problem for dynamical systems. Then, by using this inequality and the Lyapunov functional approach, quasisynchronization criteria are set for RDNNs. The desired control gain is gained from solving a group of linear matrix inequalities. Moreover, in the absence of deception attacks, the exponential synchronization problem is studied for RDNNs. In the end, simulation results are given to demonstrate the usefulness of the theoretical analysis. [ABSTRACT FROM AUTHOR]

It is known that many-objective optimization problems (MaOPs) often face the difficulty of maintaining good diversity and convergence in the search process due to the high-dimensional objective space. To address this issue, this article proposes a novel multiobjective framework for many-objective optimization (Mo4Ma), which transforms the many-objective space into multiobjective space. First, the many objectives are transformed into two indicative objectives of convergence and diversity. Second, a clustering-based sequential selection strategy is put forward in the transformed multiobjective space to guide the evolutionary search process. Specifically, the selection is circularly performed on the clustered subpopulations to maintain population diversity. In each round of selection, solutions with good performance in the transformed multiobjective space will be chosen to improve the overall convergence. The Mo4Ma is a generic framework that any type of evolutionary computation algorithm can incorporate compatibly. In this article, the differential evolution (DE) is adopted as the optimizer in the Mo4Ma framework, thus resulting in an Mo4Ma-DE algorithm. Experimental results show that the Mo4Ma-DE algorithm can obtain well-converged and widely distributed Pareto solutions along with the many-objective Pareto sets of the original MaOPs. Compared with seven state-of-the-art MaOP algorithms, the proposed Mo4Ma-DE algorithm shows strong competitiveness and general better performance. [ABSTRACT FROM AUTHOR]

This article presents a rough-to-fine evolutionary multiobjective optimization algorithm based on the decomposition for solving problems in which the solutions are initially far from the Pareto-optimal set. Subsequently, a tree is constructed by a modified $k$ -means algorithm on $N$ uniform weight vectors, and each node of the tree contains a weight vector. Each node is associated with a subproblem with the help of its weight vector. Consequently, a subproblem tree can be established. It is easy to find that the descendant subproblems are refinements of their ancestor subproblems. The proposed algorithm approaches the Pareto front (PF) by solving a few subproblems in the first few levels to obtain a rough PF and gradually refining the PF by involving the subproblems level-by-level. This strategy is highly favorable for solving problems in which the solutions are initially far from the Pareto set. Moreover, the proposed algorithm has lower time complexity. Theoretical analysis shows the complexity of dealing with a new candidate solution is $\mathcal {O}(M \log N)$ , where $M$ is the number of objectives. Empirical studies demonstrate the efficacy of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Let $\mathrm {W}\in \mathbb {C}^{n\times n}$ be a single-spiked Wishart matrix in the class $\mathrm {W}\sim \mathcal {CW}_{n}(m,\mathrm {I}_{n}+ \theta \mathrm {v}\mathrm {v}^{\dagger}) $ with $m\geq n$ , where ${\mathrm {I}}_{n}$ is the $n\times n$ identity matrix, $\mathrm {v}\in \mathbb {C}^{n\times 1}$ is an arbitrary vector with unit Euclidean norm, $\theta \geq 0$ is a non-random parameter, and $(\cdot)^{\dagger} $ represents the conjugate-transpose operator. Let u1 and ${\mathrm {u}}_{n}$ denote the eigenvectors corresponding to the smallest and the largest eigenvalues of W, respectively. This paper investigates the probability density function (p.d.f.) of the random quantity $Z_{\ell }^{(n)}=\left |{\mathrm {v}^{\dagger} \mathrm {u}_\ell }\right |^{2}\in (0,1)$ for $\ell =1,n$. In particular, we derive a finite dimensional closed-form p.d.f. for $Z_{1}^{(n)}$ which is amenable to asymptotic analysis as $m,n$ diverges with $m-n$ fixed. It turns out that, in this asymptotic regime, the scaled random variable $nZ_{1}^{(n)}$ converges in distribution to $\chi ^{2}_{2}/2(1+\theta)$ , where $\chi _{2}^{2}$ denotes a chi-squared random variable with two degrees of freedom. This reveals that u1 can be used to infer information about the spike. On the other hand, the finite dimensional p.d.f. of $Z_{n}^{(n)}$ is expressed as a double integral in which the integrand contains a determinant of a square matrix of dimension $(n-2)$. Although a simple solution to this double integral seems intractable, for special configurations of $n=2,3$ , and 4, we obtain closed-form expressions. [ABSTRACT FROM AUTHOR]

The job-shop scheduling problem (JSP) is NP hard, which has very important practical significance. Because of many uncontrollable factors, such as machine delay or human factors, it is difficult to use a single real-number to express the processing and completion time of the jobs. JSP with fuzzy processing time and completion time (FJSP) can model the scheduling more comprehensively, which benefits from the developments of fuzzy sets. Fuzzy relative entropy leads to a method that can evaluate the quality of a feasible solution following the comparison between the actual value and the ideal value (the due date). Therefore, the multiobjective FJSP can be transformed into a single-objective optimization problem and solved by a hybrid adaptive differential evolution (HADE) algorithm. The maximum completion time, the total delay time, and the total energy consumption of jobs will be considered. HADE adopts a mutation strategy based on DE-current-to-best. Its parameters (CR and F) are all made adaptive and normally distributed. The new individuals are selected according to the fitness value (FRE) obtained from a population consisting of N parents and N children in HADE. The algorithm is analyzed from different viewpoints. As the experimental results demonstrate, the performance of the HADE algorithm is better than those of some other state-of-the-art algorithms (namely, ant colony optimization, artificial bee colony, and particle swarm optimization). [ABSTRACT FROM AUTHOR]

Fall detection and prevention are crucial in elderly healthcare and humanoid robotic research as they help mitigate the damaging after-effects of falls. In this work, we have presented a deep-learning-based preimpact fall detection system (FDS) that detects a fall within 0.5 s of the fall initiation phase, thus providing a sufficient lead time of 0.5 s which is far better than the state-of-the-art. To achieve this, we have developed an automatic feature extraction methodology that can extract temporal features from all types of human fall data collected using wearable sensors. A deep neural classifier based on the ensemble of convolutional neural networks (CNNs) and long short-term memory networks (LSTMs) is trained on the extracted temporal features. The classifier has performed exceptionally well in detecting the Fall Initiation phase with a sensitivity of 99.24% and an F1-score of 98.79% for different types of falls. A sensitivity of 99.24% signifies that the model has sufficiently reduced the occurrence of false negatives, which is far more critical for an FDS. A concept of a transitional window is introduced to improve the reaction time of the FDS. We utilized two standard fall datasets, viz. SisFall and KFall, for the experimentation. Dataset fusion is employed to increase the generalizability of the system. This work can be utilized to design and develop fall detection devices for the Internet-of-Healthcare-Things applications (IoHT) and for imparting fall detection capabilities to humanoid robots and gait rehabilitation devices such as exoskeleton robots and smart prosthetic legs. [ABSTRACT FROM AUTHOR]

The developing optimization algorithms provide promising solutions for speeding up analog integrated circuit sizing. However, the optimization of complicated circuits whose solution regions are narrow remains to be a challenge. With a limited number of sampling points due to the restriction of computational resources, it is difficult for traditional algorithms to achieve satisfactory results for such circuits. To solve this problem, this article proposes a rule-guided genetic algorithm (RG-GA) for analog circuit optimization. Different from the random mutation approach in the traditional genetic algorithm (GA), the RG-GA introduces a design rule-guided mutation (RGM) mechanism which helps to find the solution region in a more straightforward fashion. Instead of handing over circuit optimization tasks to pure mathematical algorithms, the proposed method takes advantages of valuable design knowledge to improve searching efficiency. This novel algorithm is implemented and deployed to design a two-stage rail-to-rail operational amplifier (OPA), an LC voltage controlled oscillator (LC-VCO) and a four-stage OPA. Experimental results show that compared to the traditional GA method, the RG-GA achieves about 1.5 and 3.3 times speed enhancement for the two-stage rail-to-rail OPA and the LC-VCO, respectively. For the four-stage OPA, the RG-GA method can find an acceptable point within the given number of iterations while the traditional GA could not. [ABSTRACT FROM AUTHOR]

Retail sector around the world is going through a transition due to digitalization. Customers now demand convenience of online channel with instant fulfillment of offline channel. As a result, the concept of omnichannel retailing has come into existence. In omnichannel retailing, both the channels of a firm are fully or partially integrated. There are many strategies that may be developed by retailers for efficient functioning of the omnichannel retail chain. The choice of such a strategy depends upon the comparative benefits that can be derived from them. One such strategy gaining immense attention of retailers is the Buy Online Pickup in Store (BOPS) strategy. In order to understand this BOPS, we conducted a detail case study of an Indian multichannel retail chain who is considering the option of launching BOPS option in few of its retail stores. The focus is to evaluate and select a fixed number of operational retail stores to also act as pick-up points for online deliveries. K-means clustering is utilized for the formation of retail zones as per the number of retail stores to be selected. The retail stores in each zone are evaluated on the basis of a number of criteria such as demand, inventory carrying capacity, cost, and population characteristics using analytical hierarchy process and complex proportional assessment-grey. Based on these evaluations, the best suitable retail store is selected for expansion from each zone. The result implications drawn in the study can provide an understanding of the future strategies that can be developed by the decision makers of the firm for launching the BOPS option. This article can also be beneficial to other Indian retail firms who can derive meaningful insights from the findings. [ABSTRACT FROM AUTHOR]

Nonlinear equation systems (NESs) are ubiquitous, and solving them is an important yet challenging task in numerical computation. Most of the NESs usually contain multiple roots. To locate multiple roots, the use of evolutionary algorithms attracts more attention recently. For the purpose of finding different roots within a limited computational budget in a single run simultaneously, in this article, a hybrid niching-based differential evolution with two archives, referred to as HNDE/2A, is proposed. It can be featured as: 1) two niching techniques, i.e., crowding and speciation, are combined to balance the diversity and convergence; 2) a root archive is used to save the found roots during the run. Additionally, if a root is saved into the root archive, it will be reinitialized immediately to further promote the population diversity; and 3) an inferior offspring archive is presented to utilize the useful information of the inferior offspring. To evaluate the performance of our proposal, 30 problems are chosen as the test suite. Moreover, HNDE/2A is also used to solve two real-world problems. Compared with other algorithms, HNDE/2A yields better results in terms of the root ratio and the success rate with a less computational budget. [ABSTRACT FROM AUTHOR]

Colormapping is an effective and popular visualization technique for analyzing patterns in scalar fields. Scientists usually adjust a default colormap to show hidden patterns by shifting the colors in a trial-and-error process. To improve efficiency, efforts have been made to automate the colormap adjustment process based on data properties (e.g., statistical data value or histogram distribution). However, as the data properties have no direct correlation to the spatial variations, previous methods may be insufficient to reveal the dynamic range of spatial variations hidden in the data. To address the above issues, we conduct a pilot analysis with domain experts and summarize three requirements for the colormap adjustment process. Based on the requirements, we formulate colormap adjustment as an objective function, composed of a boundary term and a fidelity term, which is flexible enough to support interactive functionalities. We compare our approach with alternative methods under a quantitative measure and a qualitative user study (25 participants), based on a set of data with broad distribution diversity. We further evaluate our approach via three case studies with six domain experts. Our method is not necessarily more optimal than alternative methods of revealing patterns, but rather is an additional color adjustment option for exploring data with a dynamic range of spatial variations. [ABSTRACT FROM AUTHOR]

A mixed-integer programming (MIP) problem contains both constraints and integer restrictions. Integer restrictions divide the feasible region defined by constraints into multiple discontinuous feasible parts. In particular, the number of discontinuous feasible parts will drastically increase with the increase of the number of integer decision variables and/or the size of the candidate set of each integer decision variable. Due to the fact that the optimal solution is located in one of the discontinuous feasible parts, it is a challenging task to solve a MIP problem. This article presents a cutting and repulsion-based evolutionary framework (called CaR) to solve MIP problems. CaR includes two main strategies: 1) the cutting strategy and 2) the repulsion strategy. In the cutting strategy, an additional constraint is constructed based on the objective function value of the best individual found so far, the aim of which is to continuously cut unpromising discontinuous feasible parts. As a result, the probability of the population entering a wrong discontinuous feasible part can be decreased. In addition, in the repulsion strategy, once it has been detected that the population has converged to a discontinuous feasible part, the population will be reinitialized. Moreover, a repulsion function is designed to repulse the previously explored discontinuous feasible parts. Overall, the cutting strategy can significantly reduce the number of discontinuous feasible parts and the repulsion strategy can probe the remaining discontinuous feasible parts. Sixteen test problems developed in this article and two real-world cases are used to verify the effectiveness of CaR. The results demonstrate that CaR performs well in solving MIP problems. [ABSTRACT FROM AUTHOR]

In the field of optimization and machine learning, the statistical assessment of results has played a key role in conducting algorithmic performance comparisons. Classically, null hypothesis statistical tests have been used. However, recently, alternatives based on Bayesian statistics have shown great potential in complex scenarios, especially when quantifying the uncertainty in the comparison. In this work, we delve deep into the Bayesian statistical assessment of experimental results by proposing a framework for the analysis of several algorithms on several problems/instances. To this end, experimental results are transformed to their corresponding rankings of algorithms, assuming that these rankings have been generated by a probability distribution (defined on permutation spaces). From the set of rankings, we estimate the posterior distribution of the parameters of the studied probability models, and several inferences concerning the analysis of the results are examined. Particularly, we study questions related to the probability of having one algorithm in the first position of the ranking or the probability that two algorithms are in the same relative position in the ranking. Not limited to that, the assumptions, strengths, and weaknesses of the models in each case are studied. To help other researchers to make use of this kind of analysis, we provide a Python package and source code implementation at https://zenodo.org/record/6320599. [ABSTRACT FROM AUTHOR]

Benchmarking various metaheuristics and their new enhancements, strategies, and adaptation mechanisms has become standard in computational intelligence research. Recently, many challenges and issues regarding fair comparisons and recommendations toward good practices for benchmarking of metaheuristic algorithms, have been identified. This article is aimed at an important issues in metaheuristics design and benchmarking, which are boundary strategies or boundary control methods (BCMs). This work aims to investigate whether the choice of a BCM could significantly influence the performance of competitive algorithms. The experiments encompass the top three performing algorithms from IEEE CEC competitions 2017 and 2020 with six different BCMs. We provide extensive statistical analysis and rankings resulting in conclusions and recommendations for metaheuristics researchers and possibly also for the future direction of benchmark definitions. We conclude that the BCM should be considered another vital metaheuristics input variable for unambiguous reproducibility of results in benchmarking and for a better understanding of population dynamics, since the BCM setting could impact the optimization method performance. [ABSTRACT FROM AUTHOR]

Band selection (BS) is a widely used dimensionality reduction technique for hyperspectral images. However, most of existing evolutionary algorithms focus on searching a globally optimal band subset under a fixed size, and their obtained band subsets may still contain a large number of redundant bands. In order to simultaneously obtain multiple optimal band subsets with different sizes, this article proposes an unsupervised multitask artificial bee colony (ABC) BS algorithm based on variable-size clustering (MBBS-VC). First, a variable-size band clustering method based on worst class decomposition is developed, based on which the BS problem can be modeled as a multitask optimization problem. Next, a multitask multimicrogroup bee colony algorithm with variable coding length is proposed to simultaneously search multiple optimal band subsets with different sizes. Moreover, several new strategies, including the intergroup collaboration strategy based on bidirectional neighborhood learning and the multimeasure integration judgment (MIJ) mechanism, are designed to improve the performance of MBBS-VC. In this article, the hyperspectral BS problem is transformed into a multitask optimization problem for the first time. Finally, compared with 15 classical BS algorithms on several commonly used datasets, experimental results verify the superiority of the proposed BS algorithm. [ABSTRACT FROM AUTHOR]

Evolutionary sequential transfer optimization is a paradigm that leverages search experience from solved source optimization tasks to accelerate the evolutionary search of a target task. Even though many algorithms have been developed, they mainly focus on objective-homogeneous problems, where the source and target tasks possess a similar number of objectives. In this work, we explore objective-heterogeneous problems, in which knowledge transfers across single-objective optimization problems (SOPs), multiobjective optimization problems (MOPs), and many-objective optimization problems (MaOPs). Objective-heterogeneous problems challenge the existing methods due to the diverse search and objective spaces between the source and the target task. To address this issue, we present a decision variable analysis-based transfer method that can conduct knowledge transfer across problems with the different numbers of objectives. We first separate decision variables of MOPs and MaOPs into convergence-related variables (CVs) and diversity-related variables (DVs), according to their roles while treating variables of SOPs as CVs. Then, we propose a convergence transfer module to transfer knowledge of CVs to speed up the convergence. It aligns both solutions and fitness ranks for preserving fitness rank consistency between the source and target tasks, whereby accelerating search speed. Besides, a diversity transfer module is presented to refine the distribution of DVs to maintain the population diversity. The experimental results on objective-heterogeneous test problems and a real-world case study have demonstrated the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Finding the best configuration of algorithms’ hyperparameters for a given optimization problem is an important task in evolutionary computation. We compare in this work the results of four different hyperparameter optimization (HPO) approaches for a family of genetic algorithms (GAs) on 25 diverse pseudo-Boolean optimization (PBO) problems. More precisely, we compare previously obtained results from a grid search with those obtained from three automated configuration techniques: 1) iterated racing; 2) mixed-integer parallel-efficient global optimization (MIP-EGO); and 3) mixed-integer evolutionary strategies. Using two different cost metrics: 1) expected running time (ERT) and 2) the area under the empirical cumulative distribution function (ECDF) curve, we find that in several cases the best configurations with respect to ERT are obtained when using the area under the ECDF curve as the cost metric during the configuration process. Our results suggest that even when interested in ERT performance, it might be preferable to use anytime performance measures for the configuration task. We also observe that tuning for ERT is much more sensitive with respect to the budget that is allocated to the target algorithms. [ABSTRACT FROM AUTHOR]

This work assesses the efficacy of evolutionary algorithms (EAs) using an intuitive multidimensional scaling (MDS) visualization of the evolution of a population. We propose the use of landmark MDS (LMDS) to overcome computational challenges inherent to visualizing many-objective and complex problems with MDS. For the benchmark problems we tested, LMDS is akin to MDS visually, whilst requiring less than 1% of the time and memory necessary to produce an MDS visualization of the same objective space solutions, leading to the possibility of online visualizations for multi- and many-objective optimization evaluation. Using multi- and many-objective problems from the DTLZ and WFG benchmark test suites, we analyze how Landmark MDS visualizations can offer far greater insight into algorithm performance than using traditional algorithm performance metrics such as hypervolume alone, and can be used to complement explicit performance metrics. Ultimately, this visualization allows the visual identification of problem features and assists the decision maker in making intuitive recommendations for algorithm parameters/operators for creating and testing better EAs to solve multi- and many-objective problems. [ABSTRACT FROM AUTHOR]

Neural architecture search (NAS), which automatically designs the architectures of deep neural networks, has achieved breakthrough success over many applications in the past few years. Among different classes of NAS methods, evolutionary computation-based NAS (ENAS) methods have recently gained much attention. Unfortunately, the development of ENAS is hindered by unfair comparison between different ENAS algorithms due to different training conditions and high computational cost caused by expensive performance evaluation. This article develops a platform named BenchENAS, in short for benchmarking evolutionary NAS, to address these issues. BenchENAS makes it easy to achieve fair comparisons between different algorithms by keeping them under the same settings. To accelerate the performance evaluation in a common lab environment, BenchENAS designs a novel and generic efficient evaluation method for the population characteristics of evolutionary computation. This method has greatly improved the efficiency of the evaluation. Furthermore, BenchENAS is easy to install and highly configurable and modular, which brings benefits in good usability and easy extensibility. This article conducts efficient comparison experiments on eight ENAS algorithms with high GPU utilization on this platform. The experiments validate that the fair comparison issue does exist in the current ENAS algorithms, and BenchENAS can alleviate this issue. A Website has been built to promote BenchENAS at https://benchenas.com , where interested researchers can obtain the source code and document of BenchENAS for free. [ABSTRACT FROM AUTHOR]

Conventional multiobjective optimization algorithms (MOEAs) with or without preferences are successful in solving multi- and many-objective optimization problems. However, a strong hypothesis underlying their performance is that MOEAs are able to find a representative solution set to cover the entire Pareto-optimal front (PF) and decision makers are able to conveniently and precisely articulate their preference, which is not always easy to fulfill in practice. Accordingly, it is suggested that representative solutions in the naturally interesting regions of the PF rather than the whole PF should be targeted. A large body of research has been proposed to search or identify the knees or knee regions over the past decades. Therefore, this article aims to provide a comprehensive survey of the research on knee-oriented optimization. We start with a discussion of the importance and basic concepts of the knees, followed by a summary of knee-oriented benchmarks and indicators. After that, knee-oriented frameworks and techniques, and real-world applications are presented. Finally, potential challenges are pointed out and a few promising future lines of research are suggested. The survey offers a new perspective to develop MOEAs for solving multi- and many-objective optimization problems. [ABSTRACT FROM AUTHOR]

Distributed population dynamics offer a game-theoretical framework for distributed decision making. As such, the application of these methods to the control of networked systems has been widely studied in the literature. The theoretical analyses available in the literature have only considered continuous-time formulations of distributed population dynamics. However, given that modern computers can only process information in a discrete-time fashion, the practical implementation of distributed population dynamics-based methods is inevitably discrete. In consequence, it is paramount to extend the available theory to a more implementable discrete-time approach. For that reason, in this article, we provide a discrete-time analysis of a general class of distributed population dynamics, and we derive sufficient conditions on the discretization time to ensure the asymptotic stability of the discretized dynamics. To illustrate the relevance and performance of the proposed methods, we apply the developed theory to distributed optimization and control problems, including a real multirobot platform, which consider noncomplete communication networks and coupled constraints. [ABSTRACT FROM AUTHOR]

We study the problem of estimating the total number of searches (volume) of queries in a specific domain, which were submitted to a search engine in a given time period. Our statistical model assumes that the distribution of searches follows a Zipf’s law, and that the observed sample volumes are biased accordingly to three possible scenarios. These assumptions are consistent with empirical data, with keyword research practices, and with approximate algorithms used to take counts of query frequencies. A few estimators of the parameters of the distribution are devised and experimented, based on the nature of the empirical/simulated data. For continuous data, we recommend using nonlinear least square regression (NLS) on the top-volume queries, where the bound on the volume is obtained from the well-known Clauset, Shalizi and Newman (CSN) estimation of power-law parameters. For binned data, we propose using a Chi-square minimization approach restricted to the top-volume queries, where the bound is obtained by the binned version of the CSN method. Estimations are then derived for the total number of queries and for the total volume of the population, including statistical error bounds. We apply the methods on the domain of recipes and cooking queries searched in Italian in 2017. The observed volumes of sample queries are collected from Google Trends (continuous data) and SearchVolume (binned data). The estimated total number of queries and total volume are computed for the two cases, and the results are compared and discussed. [ABSTRACT FROM AUTHOR]

Triangle count is a critical parameter in mining relationships among people in social networks. However, directly publishing the findings obtained from triangle counts may bring potential privacy concern, which raises great challenges and opportunities for privacy-preserving triangle counting. In this paper, we choose to use differential privacy to protect triangle counting for large scale graphs. To reduce the large sensitivity caused in large graphs, we propose a novel graph projection method that can be used to obtain an upper bound for sensitivity in different distributions. In particular, we publish the triangle counts satisfying the node-differential privacy with two kinds of histograms: the triangle count distribution and the cumulative distribution. Moreover, we extend the research on privacy preserving triangle counting to one of its applications, the local clustering coefficient. Experimental results show that the cumulative distribution can fit the real statistical information better, and our proposed mechanism has achieved better accuracy for triangle counts while maintaining the requirement of differential privacy. [ABSTRACT FROM AUTHOR]

This article proposes an adaptive human pilot model that is able to mimic the crossover model in the presence of uncertainties. The proposed structure is based on the model reference adaptive control, and the adaptive laws are obtained using the Lyapunov–Krasovskii stability criteria. The model can be employed for human-in-the-loop stability and performance analyses incorporating different types of controllers and plant types. For validation purposes, an experimental setup is employed to collect data and a statistical analysis is conducted to measure the predictive power of the pilot model. [ABSTRACT FROM AUTHOR]

To fight against infectious diseases (e.g., SARS, COVID-19, Ebola, etc.), government agencies, technology companies and health institutes have launched various contact tracing approaches to identify and notify the people exposed to infection sources. However, existing tracing approaches can lead to severe privacy and security concerns, thereby preventing their secure and widespread use among communities. To tackle these problems, this paper proposes CoAvoid, an edge-based, privacy-preserved contact tracing system that features good dependability and usability. CoAvoid leverages the Google/Apple Exposure Notification (GAEN) API to achieve decent device compatibility and operating efficiency. It utilizes Bluetooth Low Energy (BLE) to detect close contact with other people and leverages GPS with fine-grained matching algorithms to verify user information. In addition, to enhance privacy protection, CoAvoid applies fuzzification and obfuscation measures to shelter sensitive data, making both servers and users agnostic to information of both low and high-risk populations. The evaluation demonstrates good efficacy and security of CoAvoid. Compared with four state-of-the-art contact tracing applications, CoAvoid can reduce the size of upload data by at least 90% and reduce the verification time by 92%. More importantly, CoAvoid can preserve user privacy and resist replay and wormhole attacks in all analysis scenarios. [ABSTRACT FROM AUTHOR]

Extracted driving behavior of human driven vehicles can benefit the development of various applications like trajectory prediction or planning, abnormal driving detection, driving behavior classification, traffic simulation modeling, etc. In this paper, we focus on modeling human driving behavior in order to find simplifications for trajectory planning. Using a time-discrete kinematic bicycle model with the vehicle’s acceleration and steering rate as inputs, we model the human driven trajectories of an urban intersection drone dataset for different input sampling times. While most planning algorithms are using input sampling times below 0.33s, we are able to model 98.2% of the human driven trajectories of the investigated dataset with a sampling time of 0.6s. Using longer input sampling times can result in smoother trajectories and longer planning horizons, and thus more efficient trajectories. In a next step, we analyze the correlations between the input of our model and the current state/last input. Such a priori knowledge could simplify common planning algorithms like model predictive control or tree-search based planners by limiting the action space of the ego-vehicle. We propose nonlinear transformations for steering rate and steering angle to represent correlations between speed, acceleration, steering angle and steering rate. In the transformed space the statistics are very well modeled by multivariate Gaussian distributions. Using a multivariate Gaussian, a fast usable behavior model is extracted which is independent of the environment. [ABSTRACT FROM AUTHOR]

The widespread existence of uncertainty makes power system control and decision-making face various risks, especially the interconnection of modern large-scale power systems, which makes its influence range more extensive. In smart grid, the renewable energy prediction and uncertainty analysis methods represented by wind power generation have attracted extensive attention. In this study, an adaptive parameter selection method based on Kullback−Leibler divergence is designed to fully extract the potential information of data. The grey wolf optimization based on reverse learning is used for prediction optimization, identifying historical data, processing dynamic information, and improving the accuracy of the model. In interval prediction, an interval prediction method combining empirical error, quantile statistics, and Markov Chain Monte Carlo is creatively proposed to generate a large number of independent error cases, and form a series of interval predictions with given confidence according to the quantile of random simulation results. Taking wind speed forecasting as an example, it is verified that the method proposed in this article has good performance. Using interval analysis to quantify the volatility and randomness can provide new ideas for load forecasting in power system and renewable energy output forecasting. [ABSTRACT FROM AUTHOR]

RRAM-based in-memory computing (IMC) effectively accelerates deep neural networks (DNNs). Furthermore, model compression techniques, such as quantization and pruning, are necessary to improve algorithm mapping and hardware performance. However, in the presence of RRAM device variations, low-precision and sparse DNNs suffer from severe post-mapping accuracy loss. To address this, in this work, we investigate a new metric, model stability, from the loss landscape to help shed light on accuracy loss under variations and model compression, which guides an algorithmic solution to maximize model stability and mitigate accuracy loss. Based on statistical data from a CMOS/RRAM 1T1R test chip at 65nm, we characterize wafer-level RRAM variations and develop a cross-layer benchmark tool that incorporates quantization, pruning, device variations, model stability, and IMC architecture parameters to assess post-mapping accuracy and hardware performance. Leveraging this tool, we show that a loss-landscape-based DNN model selection for stability effectively tolerates device variations and achieves a post-mapping accuracy higher than that with 50% lower RRAM variations. Moreover, we quantitatively interpret why model pruning increases the sensitivity to variations, while a lower-precision model has better tolerance to variations. Finally, we propose a novel variation-aware training method to improve model stability, in which there exists the most stable model for the best post-mapping accuracy of compressed DNNs. Experimental evaluation of the method shows up to 19%, 21%, and 11% post-mapping accuracy improvement for our 65nm RRAM device, across various precision and sparsity, on CIFAR-10, CIFAR-100, and SVHN datasets, respectively. [ABSTRACT FROM AUTHOR]