Dynamic Alignment Mask CTC: Improved Mask CTC With Aligned Cross Entropy

Currently, the research on the predictions of remaining useful life (RUL) of rotating machinery mainly focuses on the process of health indicator (HI) construction and the determination of the first prediction time (FPT). In complex industrial environments, the influence of environmental factors such as noise may affect the accuracy of RUL predictions. Accurately estimating the remaining useful life of bearings plays a vital role in reducing costly unscheduled maintenance and increasing machine reliability. To overcome these problems, a health indicator construction and prediction method based on multi-featured factor analysis are proposed. Compared with the existing methods, the advantages of this method are the use of factor analysis, to mine hidden common factors from multiple features, and the construction of health indicators based on the maximization of variance contribution after rotation. A dynamic window rectification method is designed to reduce and weaken the stochastic fluctuations in the health indicators. The first prediction time was determined by the cumulative gradient change in the trajectory of the HI. A regression-based adaptive prediction model is used to learn the evolutionary trend of the HI and estimate the RUL of the bearings. The experimental results of two publicly available bearing datasets show the advantages of the method.

BackgroundA new learning-based patient similarity measurement was proposed to measure patients’ similarity for heterogeneous electronic medical records (EMRs) data.MethodsWe first calculated feature-level similarities according to the features’ attributes. A domain expert provided patient similarity scores of 30 randomly selected patients. These similarity scores and feature-level similarities for 30 patients comprised the labeled sample set, which was used for the semi-supervised learning algorithm to learn the patient-level similarities for all patients. Then we used the k-nearest neighbor (kNN) classifier to predict four liver conditions. The predictive performances were compared in four different situations. We also compared the performances between personalized kNN models and other machine learning models. We assessed the predictive performances by the area under the receiver operating characteristic curve (AUC), F1-score, and cross-entropy (CE) loss.ResultsAs the size of the random training samples increased, the kNN models using the learned patient similarity to select near neighbors consistently outperformed those using the Euclidean distance to select near neighbors (all P values < 0.001). The kNN models using the learned patient similarity to identify the top k nearest neighbors from the random training samples also had a higher best-performance (AUC: 0.95 vs. 0.89, F1-score: 0.84 vs. 0.67, and CE loss: 1.22 vs. 1.82) than those using the Euclidean distance. As the size of the similar training samples increased, which composed the most similar samples determined by the learned patient similarity, the performance of kNN models using the simple Euclidean distance to select the near neighbors degraded gradually. When exchanging the role of the Euclidean distance, and the learned patient similarity in selecting the near neighbors and similar training samples, the performance of the kNN models gradually increased. These two kinds of kNN models had the same best-performance of AUC 0.95, F1-score 0.84, and CE loss 1.22. Among the four reference models, the highest AUC and F1-score were 0.94 and 0.80, separately, which were both lower than those for the simple and similarity-based kNN models.ConclusionsThis learning-based method opened an opportunity for similarity measurement based on heterogeneous EMR data and supported the secondary use of EMR data.

Optimal load clipping with time of use rates

Prediction Model of Transpiration Rate of Strawberry in Closed Cultivation Based on DBN-LSSVM Algorithm

BackgroundA critical challenge in immunotherapy research is the distinction between tumor-specific CD39+CD8+ T cells and non-specific ‘bystander’ cells.1 Deep classifiers for identifying cells in hematoxylin and eosin (H&E)-stained images are...

The applicability of the dynamic programming method to two-dimensional slope stability analyses is studied. The critical slip surface is defined as the slip surface that yields the minimum value of an optimal function. The only assumption regarding the shape of the critical slip surface is that the surface is an assemblage of linear segments. Stresses acting along the critical slip surface are computed using a finite element stress analysis. Assumptions associated with limit equilibrium methods of slices related to the shape of the critical slip surface and the relationship between interslice forces are no longer required. A computer program named DYNPROG was developed based on the proposed analytical procedure, and numerous example problems have been analyzed. Results obtained when using DYNPROG were compared with those obtained when using several well-known limit equilibrium methods. The comparisons demonstrate that the dynamic programming method provides a superior solution when compared with conventional limit equilibrium methods. Analyses conducted also show that factors of safety computed when using the dynamic programming method are generally slightly lower than those computed using conventional limit equilibrium methods of slices; however, as Poisson's ratio approaches 0.5, the computed factors of safety from the dynamic programming method and the limit equilibrium method appear to become similar.Key words: dynamic programming, slope stability, stress analysis, optimization theory, limit equilibrium methods of slices.

In Short-term Cascade Reservoirs Optimal Operation (SCROO), the flow from upstream reservoir to downstream reservoir which propagates in the natural channel can be generalized as not only transposition but also attenuation. In order to get the relative accurate inflow series of downstream reservoir, the paper tries to adopt the Muskingum model to simulate the inflow of the downstream reservoir instead of the easier processing in most papers of SCROO: ignoring the flow attenuation in natural channel. Considering the flow attenuation between the cascade reservoirs, the SCROO problem do not satisfy the “Principe of Optimality” anymore, so the Dynamic Programming(DP) is no longer applicable. The paper proposes a new improved DP named Multi-Stage Dynamic Programming (MSDP) based on DP. In MSDP, multi-stage’s outflow of upstream reservoir is taken into account at the same time and then inflow of downstream reservoir can be calculated by the Muskingum model, which can include much more flow information than the easier processing with DP, and the inflow of downstream reservoir will be closer to the actual one. The paper takes the cascade hydro-power stations consisting of Jindong and Guandi in Yalong river basin as an example to solve the SCROO problem with MSDP. By comparing the operation result of MSDP and DP with the easier processing, the operation strategy of MSDP can gain further benefits than DP’s in actual operation.

In face recognition, the class mean of the training samples may deviate from the class center in small sample size. The method based on adaptively weighted fisherface is one of the approaches to deal with the problem. However, it didnpsilat consider the face recognition efficiency. To improve recognition efficiency, the paper proposes a highly-efficient face recognition method based on weighted LDA. Firstly, the wavelet transform is applied to the face image so that the lowest resolution sub-image of the face image is obtained. Secondly, the dimension of sub-image is reduced by 2DPCA. In the end, the class means are updated by using the weighted feature vector in the reduced order subspace. The traditional LDA is improved by using the new class means. The experiments on the ORL face database show that the proposed method can achieve higher recognition rate and efficiency as well as better implementation result.

&lt;p&gt;A potentially parallel iterative algorithm for the solution of the unconstrained N-stage decision problem of Dynamic Programming is developed. This new solution method, known as Variable Metric Dynamic Programming, is based on the use of variable metric minimisation techniques to develop quadratic approximations to the optimal cost function for each stage. The algorithm is applied to various test problems, and a comparison with an existing similar algorithm proves favourable. The Variable Metric Dynamic Programming solution method is used in the implementation of an adaptive highlevel scheduling mechanism on a multiprogrammed computer in a university environment. This demonstrates a practical application of the new algorithm. More importantly, the application of Variable Metric Dynamic Programming to a scheduling problem illustrates how Mathematical Programming may be used in complex computer scheduling problems to provide in a natural way the required dynamic feedback mechanisms.&lt;/p&gt;

Dynamic programming with successive approximation has been used in the past for optimizing multi-reservoir water resources systems. In this study, a State Incremental Dynamic Programming (SIDP) model is developed for energy optimization of multi-reservoir systems. A random file access method is used to generate initial and intermediate data and cope with the curse of dimensionality of dynamic programming. The conventional dynamic programming method is used for each single reservoir to find the initial trajectory of the reservoirs. Then, the computer program developed in the study is applied to the multipurpose-multi-reservoir system in lower Seyhan basin, which comprises six reservoirs, some serial and some parallel. Extended historical flows are used to first maximize firm energy in the critical period, and then total energy over the entire period of flow records. The program is run with 50-year long segments (20 flow scenarios) of the synthetic flow data generated by using the hec-4 generalized computer program to account for the stochastic nature of streamflows. A 20% approximate increase in total energy is obtained by using the developed model for the lower seyhan basin system as compared to that calculated previously by conventional methods.

This paper mainly studies the decision-making problems encountered by enterprises in the production process, including whether to inspect parts and finished products, and how to handle unqualified parts and finished products. By applying operations research and statistical knowledge, and adopting dynamic programming and one-sided hypothesis testing methods, relevant decision-making problems are effectively solved. In this paper, the authors propose a sampling inspection method to parameterize the defect rate of parts and optimize decision-making schemes for different production stages. Finally, the study analyzes the impact of various decisions on the economic benefits of enterprises, constructs corresponding models using dynamic programming, and derives optimal solutions under various decisions. For the first problem, based on the basic principles of statistics, this paper uses one-sided hypothesis testing to detect whether the defect rate of parts exceeds the nominal value. In the case of a large sample size, the central limit theorem is cleverly applied to approximate the binomial distribution to the normal distribution, thereby simplifying the calculation process. And based on the principle of sampling inspection, detailed judgments are made on whether to accept or reject parts under different confidence levels. When studying the second problem, this paper cleverly adopts the dynamic programming method to conduct detailed decision analysis on the three key stages of the enterprise production process: part inspection, finished product inspection, and unqualified product handling. And through the reverse analysis method, starting from the final product, the decision-making of each stage is gradually optimized to ensure that the risk is minimized while controlling costs. By evaluating the relationship between inspection costs and potential losses, as well as the specific impact of different handling methods for unqualified products on the economic benefits of enterprises, the goal is to maximize economic benefits while ensuring product quality. The third problem further enhances the complexity of decision-making based on the second problem, considering multiple processes and multiple parts to optimize decision-making in the multi-stage production process. This problem still adopts the reverse analysis method of dynamic programming to construct a more complex dynamic programming model, comprehensively considering the inspection costs, unqualified product handling costs, and potential market risks of each production stage. In the process of model construction, in-depth analysis is conducted on how to effectively handle parts, semi-finished products, unqualified semi-finished products, finished products, and unqualified finished products at different production stages, including different decisions for each stage. Through careful analysis, it provides enterprises with an optimal decision-making scheme for multiple processes and multiple parts.

Bayesian network structure learning based on dynamic programming strategy can be used to find the optimal graph structure compared with approximate search methods. The traditional dynamic programming method for Bayesian network structure learning is a depth-first-based strategy, which is inefficient. We proposed two methods to solve this problem. First, the dependency constraints were used to prune the process of calculating redundancy scores. The constraints were obtained by the conditional independence test from the observed data sets. However, it was difficult to guarantee the accuracy of the constraints, which may have led to a decrease in the accuracy of the method. Second, we proposed a breadth-first-based strategy, which enhanced efficiency greatly while also ensuring global optimality. Experimental results showed that on the standard network data sets, compared with the dynamic programming based on depth-first search (DFSDP) algorithm, dynamic programming based on constraints (CBDP) could reduce the average running time by 57.10% and that dynamic programming based on breadth-first search (BFSDP) could reduce the average running time by 50.02%. On the UCI data sets, compared with DFSDP, CBDP reduced the average running time by 40.71%, and BFSDP reduced the average running time by 81.78%.

Dynamic programming is a mathematical technique for optimizing decisions in stages by breaking a problem inton stages mthat solve recursively. It display the quantity and arrangement of invemtory to maintain production and ,businness revenue efficiently when used in inventory control management. This production planning uses a dynamic programming method based on the results of advanced calculations start from stage 1 and move forward to stage 12. the dynamic programming approach in this writing is deterministic because the pattern of the number of requests for bread is known absolutely. Decisions involving a problem are simplified little by little rather than directly. This research uses linear regression with a mathematical formula that is: followed by the defect percetage formula to forecast demand for the next 12 months and apply a dynamic program to achieve the minimum total quantity.

The segmentation of breast mass is an important step not only in quantitative diagnosis but also in Computer-Aided Diagnosis (CAD) for the mammography. The objective of this study is to develop a reliable, reproducible and automatic segmentation method, which can provide the mass contours for quantitative analysis and also can be further utilized for feature extraction in CAD system. This automatic segmentation method is based on dynamic programming. Three features - edge strength, gray level and mass size were used to obtain the precise mass contour. To improve the segmentation accuracy, we introduced an adaptive mass radius determining method for each mass candidate instead of using the fixed radius for all masses to describe size and gray features. Besides, we proposed a boundary compensation method based on mean gray values in background avoiding contour departure of existing dynamic programming. In addition, we computed edge strength using local standard deviation, which can describe the edge information more clearly. In this study, we tested 120 mammography cases consisting of 178 mass regions of interests (ROIs). The cases were randomly selected from our cooperating hospitals in China. The golden standard was obtained by manually drawing the mass contour of each case according to the doctor's opinion. All the cases have also been segmented by our algorithm, and the mean overlap percentage between the proposed method and the golden standard was 0.70±0.15. The mean overlap percentage of our methods was improved 5 percent compared to that of existing dynamic programming. The experimental results showed that our improved dynamic programming method could segment most breast masses with good robustness to mass size, and it was especially effective for the mass close to image edge. However, for adhesive masses, the method needs to be further improved.

Quality function deployment (QFD) is a useful method in product design and development and its aim is to improve the quality and to better meet customers' needs. Due to cost and other resource constraints, trade-offs are always needed. Many optimization methods have been introduced into the QFD process to maximize customer satisfaction under certain constraints. However, current optimization methods sometimes cannot give practical optimal results and the data needed are hard or costly to get. To overcome these problems, this paper proposes a dynamic programming approach for the optimization problem. We first use an extended House of Quality to gather more information. Next, limited resources are allocated to the technical attributes using dynamic programming. The value of each technical attribute can be determined according to the resources allocated to them. Compared with other optimization methods, the dynamic programming method requires less information and the optimal results are more relevant. Copyright © 2005 John Wiley & Sons, Ltd.