Force evaluation is critical to ensuring the safety of cable-strut structures during service. This study employs dynamic testing to assess the internal forces resulting from cable relaxation in prestressed cable-strut...
详细信息
Force evaluation is critical to ensuring the safety of cable-strut structures during service. This study employs dynamic testing to assess the internal forces resulting from cable relaxation in prestressed cable-strut structures. A cross-model cross-mode algorithm is utilized to establish a cable force evaluation model. This approach broadens the range of available modes and addresses mismatches between modes before and after cable force loss. To enhance the accuracy and reliability of the force evaluation, a robust sparse Bayesian learning method is proposed. Measurement noise is modeled as a mixture of Gaussian distributions rather than a single Gaussian distribution, enabling a more precise representation of uncertainties in force evaluation. Furthermore, a feedback-driven error optimization process is introduced to minimize residuals through multiple linear iterations. Numerical simulations demonstrate that the proposed method achieves greater evaluation accuracy compared to existing sparse Bayesian approaches. Comparative analyses under varying noise levels reveal that the proposed method is robust and effectively reduces the impact of measurement noise.
The problem of multi-microphone blind audio source separation in noisy environment is addressed. The estimation of the acoustic signals and the associated parameters is carried out using the expectation-maximization a...
详细信息
ISBN:
(纸本)9781728166629
The problem of multi-microphone blind audio source separation in noisy environment is addressed. The estimation of the acoustic signals and the associated parameters is carried out using the expectation-maximization algorithm. Two separation algorithms are developed using either deterministic representation or stochastic Gaussian distribution for modelling the speech signals. Under the deterministic model, the speech sources are estimated in the M-step by applying in parallel multiple minimum variance distortionless response (MVDR) beamformers, while under the stochastic model, the speech signals are estimated in the E-step by applying in parallel multiple multichannel Wiener filters (MCWF). In the simulation study, we generated a large dataset of microphone signals, by convolving speech signals, with overlapping activity patterns, by measured acoustic impulse responses. It is shown that the proposed methods outperform a baseline method in terms of speech quality and intelligibility.
Transformer-based models have significantly advanced long-term time series forecasting by leveraging self- attention mechanisms to capture long-term dependencies. However, these models face high computational costs, s...
详细信息
Transformer-based models have significantly advanced long-term time series forecasting by leveraging self- attention mechanisms to capture long-term dependencies. However, these models face high computational costs, slow inference speeds, and limitations in utilizing information from longer lookback windows. Additionally, existing methods often neglect implicit spatial dependencies between variables, and struggle with semantic misalignment and insufficient diffusion of spatial information. To address these challenges, we propose DTSFormer, a D ecoupled T emporal-Spatial Diffusion Transformer designed specifically for long-term time series forecasting: (1) DTSFormer effectively integrates temporal features with implicit spatial attributes, ensuring comprehensive utilization of both temporal and spatial information. (2) DTSFormer introduce a Mix- hop Diffusion layer to effectively propagate and aggregate spatial information while preserving the original graph structure, significantly improving the accuracy of spatial information dissemination. (3) we develop a cross-diffusion attention mechanism based on the expectation-maximization algorithm, which integrates graph structure information with varying semantics under a seasonal trend decomposition framework. This approach enhances the fusion of semantic information from different graph structures and reduces computational complexity. Our extensive experiments on multiple benchmark datasets across different domains demonstrate that DTSFormer consistently achieves state-of-the-art performance in both accuracy and efficiency. These results validate DTSFormer as a robust and scalable solution for advanced long-term time series forecasting tasks.
A trajectory is a spatio-temporal data instance in which a customer or user moves between a set of discrete states while spending a certain amount of time in each state. Using trajectory data as a proxy for customers&...
详细信息
A trajectory is a spatio-temporal data instance in which a customer or user moves between a set of discrete states while spending a certain amount of time in each state. Using trajectory data as a proxy for customers' behavior and performing clustering can help to devise targeted marketing strategies. However, the censoring is often encountered due to the inability to observe the complete trajectories. In addition, the exact number of clusters is often unknown. In this work, we propose a novel mixture model-based clustering methodology to analyze the trajectory data and decipher different user segments based on their behavior. Each cluster is profiled using a semi-Markov model while considering the effect of censoring. Each entity is assigned to a cluster based on its similarity to the cluster's profile. Entity assignments and cluster profiles are simultaneously inferred using a robust expectationmaximizationalgorithm. In the simulation study, our methodology demonstrates better performance than existing methods. The effectiveness of our methodology is further corroborated using a real data set obtained from an internet music provider, where the obtained clustering results are found to be helpful in devising better marketing strategies to target users in different segments.
In this paper, we propose a novel and highly effective variational Bayesian expectationmaximization-maximization (VBEM-M) inference method for log-linear cognitive diagnostic model (CDM). In the implementation of the...
详细信息
In this paper, we propose a novel and highly effective variational Bayesian expectationmaximization-maximization (VBEM-M) inference method for log-linear cognitive diagnostic model (CDM). In the implementation of the variational Bayesian approach for the saturated log-linear CDM, the conditional variational posteriors of the parameters that need to be derived are in the same distributional family as the priors, the VBEM-M algorithm overcomes this problem. Our algorithm can directly estimate the item parameters and the latent attribute-mastery pattern simultaneously. In contrast, Yamaguchi and Okada’s (2020a) variational Bayesian algorithm requires a transformation step to obtain the item parameters for the log-linear cognitive diagnostic model (LCDM). We conducted multiple simulation studies to assess the performance of the VBEM-M algorithm in terms of parameter recovery, execution time, and convergence rate. Furthermore, we conducted a series of comparative studies on the accuracy of parameter estimation for the DINA model and the saturated LCDM, focusing on the VBEM-M, VB, expectation-maximization, and Markov chain Monte Carlo algorithms. The results indicated that our method can obtain more stable and accurate estimates, especially for the small sample sizes. Finally, we demonstrated the utility of the proposed algorithm using two real datasets.
We give convergence guarantees for estimating the coefficients of a symmetric mixture of two linear regressions by expectationmaximization (EM). In particular, we show that the empirical EM iterates converge to the t...
详细信息
We give convergence guarantees for estimating the coefficients of a symmetric mixture of two linear regressions by expectationmaximization (EM). In particular, we show that the empirical EM iterates converge to the target parameter vector at the parametric rate, provided the algorithm is initialized in an unbounded cone. In particular, if the initial guess has a sufficiently large cosine angle with the target parameter vector, a sample-splitting version of the EM algorithm converges to the true coefficient vector with high probability. Interestingly, our analysis borrows from tools used in the problem of estimating the centers of a symmetric mixture of two Gaussians by EM. We also show that the population of EM operator for mixtures of two regressions is anti-contractive from the target parameter vector if the cosine angle between the input vector and the target parameter vector is too small, thereby establishing the necessity of our conic condition. Finally, we give empirical evidence supporting this theoretical observation, which suggests that the sample-based EM algorithm may not converge to the target vector when initial guesses are drawn accordingly. Our simulation study also suggests that the EM algorithm performs well even under model misspecification (i.e., when the covariate and error distributions violate the model assumptions).
Structured covariance matrix estimation in the presence of missing-(complex) data is addressed in this paper with emphasis on radar signal processing applications. After a motivation of the study, the array model is s...
详细信息
Structured covariance matrix estimation in the presence of missing-(complex) data is addressed in this paper with emphasis on radar signal processing applications. After a motivation of the study, the array model is specified and the problem of computing the maximum likelihood estimate of a structured covariance matrix is formulated. A general procedure to optimize the observed-data likelihood function is developed resorting to the expectation-maximization algorithm. The corresponding convergence properties are thoroughly established and the rate of convergence is analyzed. The estimation technique is contextualized for two practically relevant radar problems: beamforming and detection of the number of sources. In the former case an adaptive beamformer leveraging the EM-based estimator is presented;in the latter, detection techniques generalizing the classic Akaike information criterion, minimum description length, and Hannan-Quinn information criterion, are introduced. Numerical results are finally presented to corroborate the theoretical study.
Image segmentation is a popular technique that is used for extracting information from images, which has also gained a lot of interest lately due to its importance in different scientific fields such as the medical fi...
详细信息
Image segmentation is a popular technique that is used for extracting information from images, which has also gained a lot of interest lately due to its importance in different scientific fields such as the medical field. This paper proposes a novel image segmentation technique using expectation-maximization (EM) clustering algorithm and Grasshopper Optimizer algorithm (GOA). The proposed technique and the concept of image segmentation are effectively applied on dental radiography datasets that are collected from 120 patients with an age between 6 to 60 years old. To validate the proposed technique, a comparison in terms of purity and entropy measures is conducted against K-means, X-means, EM, and Farthest First algorithms. Based on our experimental results, the proposed technique using EM and GOA achieved the best results compared to other algorithms for all three datasets in terms of entropy and purity. The best results were obtained using the second dataset, which achieved purity value of 0.7126 and entropy value of 0.3083. Further, the proposed technique also outperforms U-net and Random Forest algorithms for the selected datasets.
作者：
Ortiz-Rosario, AlexisAdeli, HojjatBuford, John A.Ohio State Univ
Dept Biomed Engn Columbus OH 43210 USA Ohio State Univ
Dept Biomed Engn Dept Biomed Informat Dept Civil & Environm Engn 470 Hitchcock Hall2070 Neil Ave Columbus OH 43210 USA Ohio State Univ
Dept Geodet Sci Dept Elect & Comp Engn Dept Neurol 470 Hitchcock Hall2070 Neil Ave Columbus OH 43210 USA Ohio State Univ
Dept Neurosci 470 Hitchcock Hall2070 Neil Ave Columbus OH 43210 USA Ohio State Univ
Sch Hlth & Rehabil Sci Div Phys Therapy 453 W 10th AveRm 516E Columbus OH 43210 USA
Researchers often rely on simple methods to identify involvement of neurons in a particular motor task. The historical approach has been to inspect large groups of neurons and subjectively separate neurons into groups...
详细信息
Researchers often rely on simple methods to identify involvement of neurons in a particular motor task. The historical approach has been to inspect large groups of neurons and subjectively separate neurons into groups based on the expertise of the investigator. In cases where neuron populations are small it is reasonable to inspect these neuronal recordings and their firing rates carefully to avoid data omissions. In this paper, a new methodology is presented for automatic objective classification of neurons recorded in association with behavioral tasks into groups. By identifying characteristics of neurons in a particular group, the investigator can then identify functional classes of neurons based on their relationship to the task. The methodology is based on integration of a multiple signal classification (MUSIC) algorithm to extract relevant features from the firing rate and an expectation-maximization Gaussian mixture algorithm (EM-GMM) to cluster the extracted features. The methodology is capable of identifying and clustering similar firing rate profiles automatically based on specific signal features. An empirical wavelet transform (EWT) was used to validate the features found in the MUSIC pseudospectrum and the resulting signal features captured by the methodology. Additionally, this methodology was used to inspect behavioral elements of neurons to physiologically validate the model. This methodology was tested using a set of data collected from awake behaving non-human primates. (C) 2016 Elsevier B.V. All rights reserved.
We propose a blind modulation classification algorithm when the channel coefficient, the noise power and the energy of the transmitted signal are unknown at the receiver. First, under each candidate modulation scheme,...
详细信息
We propose a blind modulation classification algorithm when the channel coefficient, the noise power and the energy of the transmitted signal are unknown at the receiver. First, under each candidate modulation scheme, we evaluate the unknown parameters using the iterative expectationmaximizationalgorithm. Modulation classification is then accomplished by minimizing the distance between the log-likelihood of the received data and the expected log-likelihood under each candidate modulation scheme. Results are presented from simulations in terms of detection probability vs. SNR for the class of BPSK, QPSK, 16QAM and 64QAM modulation schemes. The results show a significant improvement over QHLRT and are very close to the upper bound ALRT-UB [1].
暂无评论