Article
Computer Science, Theory & Methods
Junyi Zhang, Angelos Dassios
Summary: This paper presents a new random probability measure called the truncated Poisson-Dirichlet process. It introduces a finite approximation for the distribution of the Dirichlet process by truncating the components in descending order according to their random weights. The proposed method has a lower truncation error compared to existing stick-breaking processes.
STATISTICS AND COMPUTING
(2023)
Article
Automation & Control Systems
Kelathodi Kumaran Santhosh, Debi Prosad Dogra, Partha Pratim Roy, Bidyut Baran Chaudhuri
Summary: The paper proposes an unsupervised and nonparametric method to learn frequently used paths in road traffic, considering time dependencies of moving objects and using distance-based scene learning. This allows for quick learning of traffic scenes without manual intervention on road markings, providing a beneficial approach for designing traffic monitoring applications.
IEEE TRANSACTIONS ON CYBERNETICS
(2021)
Article
Automation & Control Systems
Marta Catalano, Pierpaolo De Blasi, Antonio Lijoi, Igor Prunster
Summary: Bayesian hierarchical models are powerful tools for learning common latent features across multiple data sources. This study establishes theoretical guarantees for recovering the true data generating process in the Hierarchical Dirichlet Process (HDP) or a generalization of the HDP. The posterior contraction rates are affected by the relationship between sample sizes.
JOURNAL OF MACHINE LEARNING RESEARCH
(2022)
Article
Mathematics, Interdisciplinary Applications
Sergio Bacallado, Stefano Favaro, Samuel Power, Lorenzo Trippa
Summary: This paper develops a perfect sampler using the Propp-Wilson algorithm to simulate the posterior distribution of the hierarchical Pitman-Yor process, and evaluates its average running time through extensive simulations. The simulations reveal a significant dependence of running time on the parameters of the model.
Article
Chemistry, Multidisciplinary
Monika Tanwar, Hyunseok Park, Nagarajan Raghavan
Summary: This study introduces a state-based diagnostic and prognostic methodology for lubricating oil degradation using the sticky hierarchical Dirichlet process-hidden Markov model (HDP-HMM). By considering multiple states in the wear-out phase of LCM data, the proposed framework improves the precision and accuracy of diagnostics and prognostics, providing guidance for maintenance decision making.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Information Systems
Koffi Eddy Ihou, Manar Amayri, Nizar Bouguila
Summary: This article introduces a Bayesian nonparametric (BNP) approach to handle model selection and topic sharing in topic models. By applying the Hierarchical Dirichlet process (HDP) to the BNP topic model, the model is able to more effectively characterize dependencies between documents and generate more robust and realistic compression algorithms.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2022)
Article
Statistics & Probability
Jung In Seo, Yongku Kim
Summary: This paper proposes a nonparametric Bayesian approach for density estimation on an open unit interval (0,1) using binomial data. The proposed method efficiently infers the smooth density defined on (0,1) through the transformation of a random variable. A blocked Gibbs sampling procedure based on the stick-breaking representation is provided for practical implementation, which avoids the use of Metropolis-Hastings transition probability.
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
(2022)
Article
Computer Science, Theory & Methods
Uttam Chauhan, Apurva Shah
Summary: In this study, the background and advancement of topic modeling techniques are explored. The various extensions and variations of topic modeling, such as across different domains, hierarchical topic modeling, word embedded topic models, and multilingual perspectives, are discussed. Implementation, evaluation techniques, comparison matrices, technical challenges, and future directions of topic modeling are also covered.
ACM COMPUTING SURVEYS
(2021)
Article
Engineering, Biomedical
Ruoshi Wen, Qiang Wang, Zhibin Li
Summary: This study improved hand movement recognition accuracy by using a small number of sEMG sequences, achieving negative lag recognition for online applications and addressing challenges such as selecting the number of hidden states. The proposed recognition method demonstrated high accuracy and potential applications in prosthetics, rehabilitation, and robot teleoperation interfaces.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2021)
Article
Computer Science, Artificial Intelligence
Jia Yu, Hongxiang Shao
Summary: This paper proposes a method called SHDP-HMM, which can automatically infer the number of hidden states from data by defining an HDP prior distribution on transition matrices. Additionally, a parameter is utilized to reduce transition probabilities among redundant states for better modeling the duration of topics. Experimental results show that this approach outperforms traditional HMM-based methods.
APPLIED INTELLIGENCE
(2022)
Article
Biology
Aakansha Gupta, Rahul Katarya
Summary: This study proposed a topic model named PAN-LDA, which integrates COVID-19 case data and news articles into standard LDA to obtain new features, and introduces these features into machine learning algorithms to enhance the prediction of time series data. Experimental results suggest that the features obtained from PAN-LDA generate more identifiable topics and provide added value to the outcomes.
COMPUTERS IN BIOLOGY AND MEDICINE
(2021)
Article
Engineering, Mechanical
Junming Ma, Nani Bai, Yi Zhou, Chengming Lan, Hui Li, B. F. Spencer
Summary: This article proposes the use of generalized hierarchical Bayesian inference for fatigue life prediction based on general multi-parameter Weibull models. The article establishes a three-layer hierarchical Bayesian structure and uses Gibbs sampling to obtain posterior samples for parameters and hyperparameters. The results show that the scatter in fatigue life prediction for the corroded specimens becomes smaller when considering informative priors for the parameters in the Weibull model.
INTERNATIONAL JOURNAL OF FATIGUE
(2022)
Article
Computer Science, Artificial Intelligence
C. Gadd, S. Wade, A. A. Shah
Summary: This Bayesian inference framework for supervised Gaussian process latent variable models introduces a method to collapse the statistical model to overcome high correlations between latent variables and hyperparameters. By using collapsed Gibbs sampling and elliptical slice sampling, the exact hyperparameter posterior and latent posterior can be explored. Compared to variational inference, this approach leads to significant improvements in predictive accuracy and uncertainty quantification, providing deeper insights into the challenges of inference in this class of models.
Article
Automation & Control Systems
George Wynne, Francois-Xavier Briol, Mark Girolami
Summary: Gaussian processes are widely used in machine learning, statistics, and applied mathematics given their flexibility in approximating functions and quantifying uncertainty. However, when the smoothness of the model and the likelihood function are misspecified, the accuracy of Gaussian process approximations can be affected, and adjusting experimental designs and choosing kernels and hyperparameters can help alleviate this issue.
JOURNAL OF MACHINE LEARNING RESEARCH
(2021)
Article
Computer Science, Theory & Methods
Antonio Canale, Riccardo Corradin, Bernardo Nipoti
Summary: Nonparametric mixture models based on the Pitman-Yor process are flexible tools for density estimation and clustering. We propose a new sampling strategy, called importance conditional sampling (ICS), which combines attractive properties of existing methods. Simulation study shows the efficiency and stability of the proposed method for different parameter specifications. The ICS approach can be naturally extended to other computationally demanding models.
STATISTICS AND COMPUTING
(2022)
Article
Computer Science, Hardware & Architecture
Abdolvahab Khalili Sadaghiani, Samad Sheikhaei
Summary: This paper proposes a novel architecture for a low-power, high-frequency PSD estimator based on the Bartlett method, using an optimized FFT processor. It utilizes the CCSSI algorithm for computing twiddle factors and has the ability to operate on short word lengths.
JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS
(2021)
Article
Computer Science, Hardware & Architecture
AbdolVahab Khalili Sadaghiani, Samad Sheikhaei, Behjat Forouzandeh
Summary: This paper proposes a novel hardware efficient low-power Welch power spectral density estimator. The presented architecture features a multiplier-less hardware and a combined coefficient selection and shift-and-add implementation to prevent multiplications. It operates as a nonparametric estimator and can work in short word lengths, providing high performance and valid output.
JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS
(2022)
Article
Computer Science, Software Engineering
AbdolVahab Khalili Sadaghiani, Samad Sheikhaei, Behjat Forouzandeh
Summary: This paper proposes a novel method for the image interpolation problem based on two-dimensional discrete wavelet transform (DWT) with the edge preserving approach. The method addresses the issues of over-smoothing and creation of spurious edges simultaneously, and offers a solution based on statistical dependencies of image sub-bands and noise behavior. The method has a multi-faceted approach for the problem, handling each 2D-DWT image sub-band differently to preserve regularity and interpolate smooth surfaces without over-smoothing, resulting in effective reduction of jaggies and annoying artifacts.
INTERNATIONAL JOURNAL OF IMAGE AND GRAPHICS
(2023)
Article
Computer Science, Artificial Intelligence
AbdolVahab Khalili Sadaghiani, Behjat Forouzandeh
Summary: This paper proposes a new discrete cosine transform (DCT) processor that utilizes a shared-resource improved coordinate rotation digital computer (CORDIC) unit to implement micro-rotation operations, reducing resource requirements and power consumption. The processor features in-order inputs and outputs, low complexity, and a distributed controller, making it capable of achieving high performance with short word lengths compared to state-of-the-art DCT processors. The proposed processor outperforms existing prominent DCT processors with limited hardware resources.
JOURNAL OF REAL-TIME IMAGE PROCESSING
(2022)
Proceedings Paper
Engineering, Electrical & Electronic
AbdolVahab Khalili Sadaghiani, Samad Sheilkhai, Behjat Forouzandeh
2020 28TH IRANIAN CONFERENCE ON ELECTRICAL ENGINEERING (ICEE)
(2020)
Proceedings Paper
Engineering, Electrical & Electronic
AbdolVahab Khalili Sadaghiani, Mohammed Ghanbari
2019 27TH IRANIAN CONFERENCE ON ELECTRICAL ENGINEERING (ICEE 2019)
(2019)
Proceedings Paper
Engineering, Electrical & Electronic
AbdolVahab Khalili Sadaghiani, Samad Sheikhai
2019 27TH IRANIAN CONFERENCE ON ELECTRICAL ENGINEERING (ICEE 2019)
(2019)