Class Of Statistical Models Research Articles

Variational Corner Transfer Matrix Renormalization Group Method for Classical Statistical Models

In the context of tensor network states, we for the first time reformulate the corner transfer matrix renormalization group (CTMRG) method into a variational bilevel optimization algorithm. The solution of the optimization problem corresponds to the fixed-point environment pursued in the conventional CTMRG method, from which the partition function of a classical statistical model, represented by an infinite tensor network, can be efficiently evaluated. The validity of this variational idea is demonstrated by the high-precision calculation of the residual entropy of the dimer model, and is further verified by investigating several typical phase transitions in classical spin models, where the obtained critical points and critical exponents all agree with the best known results in literature. Its extension to three-dimensional tensor networks or quantum lattice models is straightforward, as also discussed briefly.

Efficient Classical Simulation of Random Shallow 2D Quantum Circuits

A central question of quantum computing is determining the source of the advantage of quantum computation over classical computation. Even though simulating quantum dynamics on a classical computer is thought to require exponential overhead in the worst case, efficient simulations are known to exist in several special cases. It was widely assumed that these easy-to-simulate cases as well as any yet-undiscovered ones could be avoided by choosing a quantum circuit at random. We prove that this intuition is false by showing that certain families of constant-depth, 2D random circuits can be approximately simulated on a classical computer in time only linear in the number of qubits and gates, even though the same families are capable of universal quantum computation and are hard to exactly simulate in the worst case (under standard hardness assumptions). While our proof applies to specific random circuit families, we demonstrate numerically that typical instances of more general families of sufficiently shallow constant-depth 2D random circuits are also efficiently simulable. We propose two classical simulation algorithms. One is based on first simulating spatially local regions which are then “stitched” together via recovery maps. The other reduces the 2D simulation problem to a problem of simulating a form of 1D dynamics consisting of alternating rounds of random local unitaries and weak measurements. Similar processes have recently been the subject of an intensive research focus, which has observed that the dynamics generally undergo a phase transition from a low-entanglement (and efficient-to-simulate) regime to a high-entanglement (and inefficient-to-simulate) regime as measurement strength is varied. Via a mapping from random quantum circuits to classical statistical mechanical models, we give analytical evidence that a similar computational phase transition occurs for both of our algorithms as parameters of the circuit architecture like the local Hilbert space dimension and circuit depth are varied and, additionally, that the effective 1D dynamics corresponding to sufficiently shallow random quantum circuits falls within the efficient-to-simulate regime. Implementing the latter algorithm for the depth-3 “brickwork” architecture, for which exact simulation is hard, we find that a laptop could simulate typical instances on a 409×409 grid with a total variation distance error less than 0.01 in approximately one minute per sample, a task intractable for previously known circuit simulation algorithms. Numerical results support our analytic evidence that the algorithm is asymptotically efficient.4 MoreReceived 4 March 2021Revised 15 January 2022Accepted 14 February 2022DOI:https://doi.org/10.1103/PhysRevX.12.021021Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI.Published by the American Physical SocietyPhysics Subject Headings (PhySH)Research AreasEntanglement productionPhase transitionsQuantum computationQuantum entanglementTechniquesComputational complexityInformation theoryIsing modelMatrix product statesTensor network methodsQuantum InformationStatistical Physics

U19-Net: a deep learning approach for obstacle detection in self-driving cars

Development of self-driving cars aims to drive safely from one point to another in a coordinated system where an on-board system should react and possibly alert drivers about the driving environments and possible collisions that may arise between drivers and obstacles. There are many deep learning approaches available for obstacle detection especially convolutional neural networks (CNNs) with improvement accuracy, and encoder–decoder networks are CNNs with a current attraction for researchers mainly because these models provide better results than classical statistical models for image segmentation and object classification tasks. This work proposes U19-Net an encoder–decoder deep model that explores the deep layers of a VGG19 model as an encoder following a symmetrical approach with an U-Net decoder designed for pixel-wise classifications. The U19-Net has end-to-end learning successfully effectiveness for the vehicle and pedestrian detection within the open-source Udacity dataset showing an IoU score of 87.08 and 78.18%, respectively. The proposed U19-Net is compared with five recent CNN networks using the AP metric, obtaining near results (less than 5%) for the faster R-CNN, one of the most commonly used networks for object detection.

Universal Behavior beyond Multifractality of Wave Functions at Measurement-Induced Phase Transitions.

We investigate the structure of many-body wave functions of 1D quantum circuits with local measurements employing the participation entropies. The leading term in system size dependence of participation entropy indicates a model-dependent multifractal scaling of the wave functions at any nonzero measurement rate. The subleading term contains universal information about measurement-induced phase transitions and plays the role of an order parameter, being constant nonzero in the error-correcting phase and vanishing in the quantum Zeno phase. We provide robust numerical evidence investigating a variety of quantum many-body systems and provide an analytical interpretation of this behavior expressing the participation entropy in terms of partition functions of classical statistical models in 2D.

Random Quantum Circuits Anticoncentrate in Log Depth

We consider quantum circuits consisting of randomly chosen two-local gates and study the number of gates needed for the distribution over measurement outcomes for typical circuit instances to be anti-concentrated, roughly meaning that the probability mass is not too concentrated on a small number of measurement outcomes. Understanding the conditions for anti-concentration is important for determining which quantum circuits are difficult to simulate classically, as anti-concentration has been in some cases an ingredient of mathematical arguments that simulation is hard and in other cases a necessary condition for easy simulation. Our definition of anti-concentration is that the expected collision probability, that is, the probability that two independently drawn outcomes will agree, is only a constant factor larger than if the distribution were uniform. We show that when the 2-local gates are each drawn from the Haar measure (or any two-design), at least $\Omega(n \log(n))$ gates (and thus $\Omega(\log(n))$ circuit depth) are needed for this condition to be met on an $n$ qudit circuit. In both the case where the gates are nearest-neighbor on a 1D ring and the case where gates are long-range, we show $O(n \log(n))$ gates are also sufficient, and we precisely compute the optimal constant prefactor for the $n \log(n)$. The technique we employ relies upon a mapping from the expected collision probability to the partition function of an Ising-like classical statistical mechanical model, which we manage to bound using stochastic and combinatorial techniques.

Network security situation awareness forecasting based on statistical approach and neural networks

AbstractThe usage of new and progressive technologies brings with it new types of security threats and security incidents. Their number is constantly growing.The current trend is to move from reactive to proactive activities. For this reason, the organization should be aware of the current security situation, including the forecasting of the future state. The main goal of organizations, especially their security operation centres, is to handle events, identify potential security incidents, and effectively forecast the network security situation awareness (NSSA). In this paper, we focus on increasing the efficiency of utilization of this part of cybersecurity. The paper’s main aim is to compare selected statistical models and models based on neural networks to find out which models are more suitable for NSSA forecasting. Based on the analysis provided in this paper, neural network methods prove a more accurate alternative than classical statistical prediction models in NSSA forecasting. In addition, the paper analyses the selection criteria and suitability of time series, which do not only reflect information about the total number of security events but represent a category of security event (e.g. recon scanning), port or protocol.

Machine learning-based prediction of postpartum hemorrhage after vaginal delivery: combining bleeding high risk factors and uterine contraction curve.

This work used a machine learning model to improve the accuracy of predicting postpartum hemorrhage in vaginal delivery. Among the 25,098 deliveries in the obstetrics department of the First Hospital of Jinan University recorded from 2016 to 2020, 10,520 were vaginal deliveries with complete study data. Further review selected 850 cases of postpartum hemorrhage (amount of bleeding > 500mL) and 54 cases of severe postpartum hemorrhage (amount of bleeding > 1000mL). Indicators of clinical risk factors for postpartum hemorrhage were retrieved from electronic medical records. Features of the uterine contraction curve were extracted 2h prior to vaginal delivery and modeled using a 49-variable machine learning with 90% of study cases used in the training set and 10% of study cases used in the test set. Accuracy was compared among the assessment table, classical statistical models, and machine learning models used to predict postpartum hemorrhage to assess their clinical use. The assessment table contained 16 high-risk factor scores to predict postpartum hemorrhage. The classical statistical model used was Logistic Regression (LR). The machine learning models were Random Forest (RF), K Nearest Neighbor (KNN), and the one integrated with Lightgbm (LGB) and LR. The effect of model prediction was evaluated by area under the receiver operating characteristic curve (AUC), namely, C-static, calibration curve Brier score, decision curve, F-measure, sensitivity (SE), and specificity (SP). 1:Among the tested tools, the machine learning model LGB + LR has the best performance in predicting postpartum hemorrhage. Its Brier, AUC, and F-measure scores are better than those of other models in each group, and its SE and SP reach 0.694 and 0.800, respectively. The predictive performance of the classical statistical model LR is AUC: 0.729, 95%CI [0.702-0.756]). 2: Verification on the testing set reveals that the features of uterine contraction contribute to the improved accuracy of the model prediction. 3: LGB + LR model suggested that among the 49 indicators for predicting severe postpartum hemorrhage, the importance of the first 10 characteristics in descending order is as follows: hematocrit (%), shock index, frequency of contractions (min-1), white blood cell count, gestational hypertension, neonatal weight (kg), time of second labor (min), mean area of contractions (mmHgs), total amniotic fluid (mL), and body mass index (BMI). The prediction effect is close to that of the model after training with all 49 features. The predictive effect was close to that of the model after training using all 49 features. 4: Contraction frequency and intensity Mean_Area (representing effective contractions) have a high predictive value for severe postpartum hemorrhage. 5: Blood loss amount within 2h has a high warning effect on postpartum hemorrhage, and the increase in AUC to 0.95 indicates that postpartum bleeding mostly occurs within 2h after delivery. Machine learning models incorporated with uterine contraction features can further improve the accuracy of postpartum hemorrhage prediction in vaginal delivery and provide a reference for clinicians to intervene early and reduce adverse pregnancy outcomes.

Insights Into the Mn(VII) and Cr(VI) Adsorption Mechanisms on Purified Diatomite/MCM-41 Composite: Experimental Study and Statistical Physics Analysis.

In this study, a purified diatomite (PD) with a concentration of diatom frustules more than 92% SiO2 was utilized to synthesize a composite of MCM-41 silica under hydrothermal conditions. The as-synthesized PD/MCM-41 composite was characterized and tested as an adsorbent for the removal of Cr(VI) and Mn(VII) ions from aqueous solution. Results of X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and Fourier-transform infrared spectroscopy (FTIR) revealed that the diatom frustules of the PD were coated with MCM-41 mesoporous silica. Experimental isotherms of Cr(VI) and Mn(VII) adsorption were fitted to classical and advanced statistical physics models at 25°C–55°C and pH 3. The Langmuir model estimated monolayer adsorption capacities ranging from 144.1 to 162.2 mg/g for Cr(VI) and 166.2 to 177.0 mg/g for Mn(VII), which improved with increasing the solution temperature. Steric and energetic parameters obtained from a monolayer adsorption model with one adsorption site was utilized to explain the adsorption mechanism at a microscopic level. The number of Cr(VI) and Mn(VII) ions adsorbed on PD/MCM-41 active site (n) were 1.25–1.27 for Cr(VI) and 1.27–1.32 for Mn(VII), thus suggesting multi-interaction mechanisms. The density of PD/MCM-41 active sites (D M) was a key parameter to explain the adsorption of these heavy metals. The adsorbed quantities were maximum at 55°C, thus obtaining 102.8 and 110.7 mg/g for Cr(VI) and Mn(VII), respectively. Cr(VI) and Mn(VII) adsorption energies ranged from 18.48 to 26.70 kJ/mol and corresponded to an endothermic adsorption with physical forces. Entropy, free enthalpy, and internal energy associated to the adsorption of Cr(VI) and Mn(VII) ions were calculated, thus indicating that the removal of these pollutants was spontaneous. Overall, this article offers new interpretations for the Cr(VI) and Mn(VII) adsorption mechanisms on PD/MCM-41 composite, which are relevant to contribute to the development of effective water treatment processes.

Feature importance: Opening a soil-transmitted helminth machine learning model via SHAP.

In the field of landscape epidemiology, the contribution of machine learning (ML) to modeling of epidemiological risk scenarios presents itself as a good alternative. This study aims to break with the ”black box” paradigm that underlies the application of automatic learning techniques by using SHAP to determine the contribution of each variable in ML models applied to geospatial health, using the prevalence of hookworms, intestinal parasites, in Ethiopia, where they are widely distributed; the country bears the third-highest burden of hookworm in Sub-Saharan Africa. XGBoost software was used, a very popular ML model, to fit and analyze the data. The Python SHAP library was used to understand the importance in the trained model, of the variables for predictions. The description of the contribution of these variables on a particular prediction was obtained, using different types of plot methods. The results show that the ML models are superior to the classical statistical models; not only demonstrating similar results but also explaining, by using the SHAP package, the influence and interactions between the variables in the generated models. This analysis provides information to help understand the epidemiological problem presented and provides a tool for similar studies.

Joint kinematics alone can distinguish hip or knee osteoarthritis patients from asymptomatic controls with high accuracy.

Osteoarthritis (OA) is one of the leading musculoskeletal disabilities worldwide, and several interventions intend to change the gait pattern in OA patients to more healthy patterns. However, an accessible way to follow up the biomechanical changes in a clinical setting is still missing. Therefore, this study aims to evaluate whether we can use biomechanical data collected from a specific activity of daily living to help distinguish hip OA patients from controls and knee OA patients from controls using features that potentially could be measured in a clinical setting. To achieve this goal, we considered three different classes of statistical models with different levels of data complexity. Class 1 is kinematics based only (clinically applicable), class 2 includes joint kinetics (semi-applicable under the condition of access to a force plate or prediction models), and class 3 uses data from advanced musculoskeletal modeling (not clinically applicable). We used a machine learning pipeline to determine which classification model was best. We found 100% classification accuracy for KneeOA-vs-Asymptomatic and 93.9% for HipOA-vs-Asymptomatic using seven features derived from the lumbar spine and hip kinematics collected during ascending stairs. These results indicate that kinematical data alone can distinguish hip or knee OA patients from asymptomatic controls. However, to enable clinical use, we need to validate if the classifier also works with sensor-based kinematical data and whether the probabilistic outcome of the logistic regression model can be used in the follow-up of patients with OA.

Improving forecasting by subsampling seasonal time series

ABSTRACT Time series forecasting plays an increasingly important role in modern business decisions. In today's data-rich environment, people often aim to choose the optimal forecasting model for their data. However, identifying the optimal model requires professional knowledge and experience, making accurate forecasting a challenging task. To mitigate the importance of model selection, we propose a simple and reliable algorithm to improve the forecasting performance. Specifically, we construct multiple time series with different sub-seasons from the original time series. These derived series highlight different sub-seasonal patterns of the original series, making it possible for the forecasting methods to capture diverse patterns and components of the data. Subsequently, we produce forecasts for these multiple series separately with classical statistical models (ETS or ARIMA). Finally, the forecasts are combined. We evaluate our approach on widely used forecasting competition data sets (M1, M3, and M4) in terms of both point forecasts and prediction intervals. We observe performance improvements compared with the benchmarks. Our approach is particularly suitable and robust for the data with higher frequency. To demonstrate the practical value of our proposition, we showcase the performance improvements from our approach on hourly load data that exhibit multiple seasonal patterns.

Estimated Renal Metabolomics at Reperfusion Predicts One-Year Kidney Graft Function.

Renal transplantation is the gold-standard procedure for end-stage renal disease patients, improving quality of life and life expectancy. Despite continuous advancement in the management of post-transplant complications, progress is still needed to increase the graft lifespan. Early identification of patients at risk of rapid graft failure is critical to optimize their management and slow the progression of the disease. In 42 kidney grafts undergoing protocol biopsies at reperfusion, we estimated the renal metabolome from RNAseq data. The estimated metabolites’ abundance was further used to predict the renal function within the first year of transplantation through a random forest machine learning algorithm. Using repeated K-fold cross-validation we first built and then tuned our model on a training dataset. The optimal model accurately predicted the one-year eGFR, with an out-of-bag root mean square root error (RMSE) that was 11.8 ± 7.2 mL/min/1.73 m2. The performance was similar in the test dataset, with a RMSE of 12.2 ± 3.2 mL/min/1.73 m2. This model outperformed classic statistical models. Reperfusion renal metabolome may be used to predict renal function one year after allograft kidney recipients.

Water Erosion Monitoring and Prediction in Response to the Effects of Climate Change Using RUSLE and SWAT Equations: Case of R’Dom Watershed in Morocco

Soil erosion is an increasingly issue worldwide, due to several factors including climate variations and humans’ activities, especially in Mediterranean ecosystems. Therefore, the aim of this paper is: (i) to quantify and to predict soil erosion rate for the baseline period (2000–2013) and a future period (2014–2027), using the Revised Universal Soil Loss Equation (RUSLE) and the Soil and Water Assessment Tool (SWAT) model in the R’Dom watershed in Morocco, based on the opportunities of Remote Sensing (RS) techniques and Geographical Information System (GIS) geospatial tools. (ii) we based on classical statistical downscaling model (SDSM) for rainfall prediction. Due to the lack of field data, the model results are validated by expert knowledge. As a result of this study, it is found that both agricultural lands and bare lands are most affected by soil erosion. Moreover, it is showed that soil erosion in the watershed was dominated by very low and low erosion. Although the area of very low erosion and low erosion continued to decrease. Hence, we hereby envisage that our contribution will provide a more complete understanding of the soil degradation in this study area and the results of this research could be a crucial reference in soil erosion studies and also may serve as a valuable guidance for watershed management strategies.

Machine learning classifiers do not improve prediction of hospitalization > 2 days after fast-track hip and knee arthroplasty compared with a classical statistical risk model.

Background and purposePrediction of postoperative outcomes and length of hospital stay (LOS) of patients is vital for allocation of healthcare resources. We investigated the performance of prediction models based on machinelearning algorithms compared with a previous risk stratification model using traditional multiple logistic regression, for predicting the risk of a LOS of > 2 days after fast-track total hip and knee replacement.Patients and methods3 different machine learning classifiers were trained on data from the Lundbeck Centre for Fast-track Hip and Knee Replacement Database (LCDB) collected from 9,512 patients between 2016 and 2017. The chosen classifiers were a random forest classifier (RF), a support vector machine classifier with a polynomial kernel (SVM), and a multinomial Naïve-Bayes classifier (NB).ResultsComparing performance measures of the classifiers with the traditional model revealed that all the models had a similar performance in terms of F1 score, accuracy, sensitivity, specificity, area under the receiver operating curve (AUC), and area under the precision-recall curve (AUPRC). A feature importance analysis of the RF classifier found hospital, age, use of walking aid, living alone, and joint operated on to be the most relevant input features. None of the classifiers reached a clinically relevant performance with the input data from the LCDB.InterpretationDespite the promising prospects of machine-learning practices for disease and risk prediction, none of the machine learning models tested outperformed the traditional multiple regression model in predicting which patients in this cohort had a LOS > 2 days.

Bayesian learning of Gaussian mixture model for calculating debris flow exceedance probability

ABSTRACT Probabilistic modelling of debris flow data provides useful information for quantitative risk assessment, such as exceedance probabilities (EPs) of debris flow quantities. This task can defy many classical statistical models because debris flow data are frequently collected over years or even decades, and the nonuniformity of the nature and the complex physical mechanism of debris flows lead to multimodal distribution characteristics of observational data. This paper proposes a Bayesian framework for learning Gaussian mixture model (GMM) of debris flow quantities (e.g. total discharge Q total and maximum impact pressure P max) and calculating their EPs for risk-informed decision making. GMM provides great flexibility to fit observation data, but are intrinsically unidentifiable due to the label switching. These computational difficulties are addressed using Random Gibbs Sampling and Bridge Sampling in the proposed framework, allowing incorporating the statistical uncertainty in GMM parameters into EP estimation. Equations are derived for the proposed approach and are illustrated using Q total and P max data at Jiangjia Ravine, China. Results show that the proposed approach identifies a bivariate GMM of Q total and P max reflecting the multimodal characteristics of the observed data and quantifies the statistical uncertainty of GMM parameters. Incorporating the statistical uncertainty into EP estimation provides robust estimates.

Inter-Purchase Time Prediction Based on Deep Learning

Inter-purchase time is a critical factor for predicting customer churn. Improving the prediction accuracy can exploit consumer’s preference and allow businesses to learn about product or pricing plan weak points, operation issues, as well as customer expectations to proactively reduce reasons for churn. Although remarkable progress has been made, classic statistical models are difficult to capture behavioral characteristics in transaction data because transaction data are dependent and short-, medium-, and long-term data are likely to interfere with each other sequentially. Different from literature, this study proposed a hybrid inter-purchase time prediction model for customers of on-line retailers. Moreover, the analysis of differences in the purchase behavior of customers has been particularly highlighted. The integrated self-organizing map and Recurrent Neural Network technique is proposed to not only address the problem of purchase behavior but also improve the prediction accuracy of inter-purchase time. The permutation importance method was used to identify crucial variables in the prediction model and to interpret customer purchase behavior. The performance of the proposed method is evaluated by comparing the prediction with the results of three competing approaches on the transaction data provided by a leading e-retailer in Taiwan. This study provides a valuable reference for marketing professionals to better understand and develop strategies to attract customers to shorten their inter-purchase times.

Optimizing Lstm and Bi-Lstm Models for Multivariate Time Series Prediction and Performance Comparison with Classical Statistical Models

Optimizing Lstm and Bi-Lstm Models for Multivariate Time Series Prediction and Performance Comparison with Classical Statistical Models

Price Prediction of Seasonal Items Using Machine Learning and Statistical Methods

Price prediction of goods is a vital point of research due to how common e-commerce platforms are. There are several efforts conducted to forecast the price of items using classic machine learning algorithms and statistical models. These models can predict prices of various financial instruments, e.g., gold, oil, cryptocurrencies, stocks, and second-hand items. Despite these efforts, the literature has no model for predicting the prices of seasonal goods (e.g., Christmas gifts). In this context, we framed the task of seasonal goods price prediction as a regression problem. First, we utilized a real online trailer dataset of Christmas gifts and then we proposed several machine learning-based models and one statistical-based model to predict the prices of these seasonal products. Second, we utilized a real-life dataset of Christmas gifts for the prediction task. Then, we proposed support vector regressor (SVR), linear regression, random forest, and ridge models as machine learning models for price prediction. Next, we proposed an autoregressive-integrated-moving-average (ARIMA) model for the same purpose as a statistical-based model. Finally, we evaluated the performance of the proposed models; the comparison shows that the best performing model was the random forest model, followed by the ARIMA model.

Dense Sampling of Time Series for Forecasting

A time series contain a large amount of information suitable for forecasting. Classical statistical and recent deep learning models have been widely used in a variety of forecasting applications. During the training data preparation stage, most models collect samples by sliding a fixed-sized window over the time axis of the input time series. We refer to this conventional method as “sparse sampling” because it cannot extract sufficient samples because it ignores another important axis representing the window size. In this study, a dense sampling method is proposed that extends the sampling space from one to two dimensions. The new space consists of time and window axes. Dense sampling provides several desirable effects, such as a larger training dataset, an intra-model ensemble, model-agnosticism, and an easier setting of the optimal window. The experiments were conducted using four real datasets: Bitcoin price, influenza-like illness, household electric power consumption, and wind speed. The mean absolute percentage error was measured extensively in terms of varying window sizes, horizons, and lengths of time series. The resulting data showed that dense sampling significantly and consistently outperformed sparse sampling. The source codes and datasets are available at https://github.com/isoh24/Dense-sampling-time-series.

Dynamic Network Slice Scaling Assisted by Attention-Based Prediction in 5G Core Network

Network slicing is a key technology in fifth-generation (5G) networks that allows network operators to create multiple logical networks over a shared physical infrastructure to meet the requirements of diverse use cases. Among core functions to implement network slicing, resource management and scaling are difficult challenges. Network operators must ensure the Service Level Agreement (SLA) requirements for latency, bandwidth, resources, etc for each network slice while utilizing the limited resources efficiently, i.e., optimal resource assignment and dynamic resource scaling for each network slice. Existing resource scaling approaches can be classified into reactive and proactive types. The former makes a resource scaling decision when the resource usage of virtual network functions (VNFs) exceeds a predefined threshold, and the latter forecasts the future resource usage of VNFs in network slices by utilizing classical statistical models or deep learning models. However, both have a trade-off between assurance and efficiency. For instance, the lower threshold in the reactive approach or more marginal prediction in the proactive approach can meet the requirements more certainly, but it may cause unnecessary resource wastage. To overcome the trade-off, we first propose a novel and efficient proactive resource forecasting algorithm. The proposed algorithm introduces an attention-based encoder-decoder model for multivariate time series forecasting to achieve high short-term and long-term prediction accuracies. It helps network slices be scaled up and down effectively and reduces the costs of SLA violations and resource overprovisioning. Using the attention mechanism, the model attends to every hidden state of the sequential input at every time step to select the most important time steps affecting the prediction results. We also designed an automated resource configuration mechanism responsible for monitoring resources and automatically adding or removing VNF instances of network slices, which helps network operators satisfy service requirements even when the traffic of end-user requests changes dynamically. Comprehensive experiments demonstrate that our proposed solution outperforms other solutions in terms of short-term and long-term predictions while reducing the cost of SLA violations and resource overprovisioning and enhancing the delay quality of network slices.