Article
Neurosciences
Danish Shaikh
Summary: This article explores the process of multisensory integration and its development in a multisensory environment. It proposes that experience dependent crossmodal synaptic plasticity may be a mechanism underlying the development of multisensory cue integration. The hypothesis is tested using a computational model and simulated experiments.
FRONTIERS IN NEURAL CIRCUITS
(2022)
Article
Behavioral Sciences
Shir Shalom-Sperber, Aihua Chen, Adam Zaidel
Summary: This study investigated cross-sensory (visual-vestibular) adaptation of self-motion perception. The results showed that even several short-duration stimuli can lead to functional adaptation of perception, suggesting that the brain monitors and adapts to supra-modal statistics of events in the environment.
Article
Multidisciplinary Sciences
Ye Lin, Sean B. Andersson
Summary: Single Particle Tracking (SPT) is a well-known tool for studying the dynamics of biological macromolecules inside living cells. The study focuses on the problem of localization and parameter estimation and proposes an Expectation Maximization (EM) based framework for simultaneous handling. Two representative methods, namely SMC-EM and U-EM, demonstrate better performance compared to standard techniques, especially at low signal levels.
Article
Automation & Control Systems
Yanling Chang, Alfredo Garcia, Zhide Wang, Lu Sun
Summary: This article discusses the (inverse) structural estimation of POMDPs based on observable sequences and implemented actions. The structural properties of an entropy regularized POMDP are analyzed, and conditions for model identifiability without knowledge of state dynamics are specified. A soft policy gradient algorithm is used to compute a maximum likelihood estimator, and an equipment replacement problem is used as an illustration.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2023)
Article
Ecology
Erin Clancey, Timothy R. Johnson, Luke J. Harmon, Paul A. Hohenlohe
Summary: The study develops a novel maximum likelihood-based method to quantify and infer the strength and form of mate preference, bridging the gap between mathematical models and empirical data. The method accurately infers model parameters and provides additional insight into the role of mate preference in various plant and animal taxa that exhibit mating preferences, facilitating the testing of evolutionary hypotheses.
Article
Economics
Angelo Mele, Lingjiong Zhu
Summary: We have developed approximate estimation methods for exponential random graph models (ERGMs), where the likelihood is proportional to an intractable normalizing constant. Instead of using Monte Carlo simulations, which can converge slowly, we propose a deterministic method based on a variational mean-field approximation. Our method provides lower and upper bounds for the approximation error, allowing us to estimate the distance between true likelihood and mean-field likelihood. Monte Carlo simulations indicate that our deterministic method performs well in practice, surpassing the limitations of our theoretical approximation bounds.
REVIEW OF ECONOMICS AND STATISTICS
(2023)
Article
Computer Science, Interdisciplinary Applications
John Paul Helveston
Summary: This paper introduces the logitr R package, which allows for fast maximum likelihood estimation of multinomial logit and mixed logit models with unobserved heterogeneity across individuals by modeling parameters that vary randomly over individuals according to a chosen distribution. It is faster than other similar packages and supports utility models specified with preference space or willingness-to-pay (WTP) space parameterizations, allowing for direct estimation of marginal WTP. The paper discusses the implications of each utility parameterization for WTP estimates and highlights design features that enable logitr's performant estimation speed, including benchmarking with similar packages and additional features designed specifically for WTP space models.
JOURNAL OF STATISTICAL SOFTWARE
(2023)
Article
Mathematics
Chanseok Park
Summary: This paper considers parameter estimation of the Weibull distribution with interval-censored data using the expectation-maximization (EM) algorithm. The results show that the estimates obtained using the EM method have superior convergence properties compared to conventional Newton-type methods. Finally, a numerical study is provided to illustrate the advantages of the proposed method.
Article
Quantum Science & Technology
Alfred Godley, Madalin Guta
Summary: In this paper, the problem of extracting maximum information from continuous-time measurements in quantum systems is addressed. The authors propose an efficient algorithm for optimal estimation of one-dimensional dynamical parameters in discrete-time input-output quantum Markov chains. The algorithm involves updating a "measurement filter" operator and determining measurement bases for the output units. A key component of the algorithm is the use of a coherent quantum absorber to "post-process" the output. The scheme offers potential for optimal continuous-time adaptive measurements, but practical implementations require further research.
Article
Computer Science, Information Systems
Lei Li, Zhiyuan Liu, Zan Zhang, Huanhuan Chen, Xindong Wu
Summary: The preference completion method combines the partial rankings of the target agent and the partial rankings from third parties to infer the agent's personalized preference over all alternatives. It utilizes weighted preference graph and maximum likelihood estimation (MLE) to settle disagreements and obtain the completed preference. To efficiently locate the edges with the minimum weight in a big graph, optimal MLE algorithm and three greedy MLE algorithms are proposed.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2023)
Article
Environmental Sciences
Qin Jiang, Yifei Dong, Jiangtao Peng, Mei Yan, Yi Sun
Summary: The paper introduces a robust MLE-based NMF model for hyperspectral unmixing, which shows superior performance compared to existing NMF methods in experiments using simulated and real hyperspectral data sets.
Article
Physics, Fluids & Plasmas
Jonathan Larson, Jukka-Pekka Onnela
Summary: Mechanistic network models allow researchers to study complex systems, but it is challenging to estimate the parameters for graphs generated with these models. This paper proposes a method of treating the node sequence as an additional parameter or a missing random variable in growing network models and maximizing the resulting likelihood. The proposed framework is tested on simulated graphs and applied to protein-protein interaction networks.
Article
Automation & Control Systems
Mohammad S. Ramadan, Robert R. Bitmead
Summary: A Maximum Likelihood recursive state estimator is proposed for non-linear state-space models, which combines a particle filter and the Expectation Maximization algorithm. Algorithms for maximum likelihood state filtering, prediction, and smoothing are derived, and their convergence properties are examined, demonstrating the effectiveness of the method in nonlinear systems.
Article
Mathematics, Applied
Mohamed Kayid, Mashael A. Alshehri
Summary: The inverse Weibull model is a simple and flexible model widely used in survival analysis, reliability theory, and other scientific fields. This study proposes a modified version of the maximum likelihood estimator to address the bias issue in the estimation of the model parameters. By defining an unbiased likelihood equation, the new estimator demonstrates a smaller mean square error in simulation results. Additionally, the proposed estimator is compared with the maximum likelihood estimator based on the analysis of three real data sets, validating its effectiveness.
Article
Mathematics
Neveka M. Olmos, Emilio Gomez-Deniz, Osvaldo Venegas
Summary: This paper studies a distribution with heavy right tail, examining its properties and tail behavior. Using the maximum likelihood method and Monte Carlo evaluation, the parameters are estimated and the performance of the estimators is evaluated. Analysis of simulated and real data demonstrates that the studied distribution can be used to model income data.
