Editorial Material
Economics
Mark Bognanni
Summary: This article introduces a method for fully Bayesian inference in the VAR-SV model and compares the different effects of using the triangular algorithm and the systemwide algorithm in the MCMC algorithm.
JOURNAL OF ECONOMETRICS
(2022)
Article
Statistics & Probability
Jacob Vorstrup Goldman, Torben Sell, Sumeetpal Sidhu Singh
Summary: The use of nondifferentiable priors in Bayesian statistics is increasing, with current methods being approximate and utilizing gradient-based diffusion for sampling. The Moreau-Yosida approximation error is characterized and a new implementation using underdamped Langevin dynamics is proposed. Piecewise-deterministic Markov processes (PDMP) can be used for exact posterior inference in cases of almost everywhere differentiability, offering a broader scope of application compared to diffusion-based methods.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2022)
Article
Mathematics
Tzong-Ru Tsai, Yuhlong Lio, Hua Xin, Hoang Pham
Summary: This study proposed a method to infer the baseline survival function using maximum likelihood estimation and Bayesian estimation methods, with a mixture distribution to characterize changes in lifetime distribution. A Markov chain Monte Carlo approach was used to overcome computational complexity, showing that the Bayesian estimation method outperformed maximum likelihood estimation in obtaining reliable estimates.
Article
Geochemistry & Geophysics
Hongyi Liu, Youkang Lu, Zebin Wu, Qian Du, Jocelyn Chanussot, Zhihui Wei
Summary: In this article, a Bayesian unmixing model considering spectral variability is proposed, which develops composite prior distributions to model the variability of abundance and endmembers. The experiments on synthetic and real data sets demonstrate the effectiveness of the proposed approach in terms of accuracy in estimating abundance, endmembers, and their variability.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Computer Science, Artificial Intelligence
Matthew Holden, Marcelo Pereyra, Konstantinos C. Zygalakis
Summary: This paper proposes a new methodology for Bayesian inference in imaging inverse problems using data-driven priors. The methodology learns the prior distribution from training data using generative models and provides rigorous underpinning for Bayesian estimators and uncertainty quantification analyses. The paper also introduces a model misspecification test and a method to identify the dimension of the latent space from training data. Experimental results show the effectiveness of the proposed approach and compare it with other data-driven regularization methods.
SIAM JOURNAL ON IMAGING SCIENCES
(2022)
Article
Engineering, Multidisciplinary
Dhruv Patel, Deep Ray, Assad A. Oberai
Summary: Inverse problems are common in various fields of science and engineering, and Bayesian inference provides a principled approach to overcome their ill-posed nature. This work presents a novel method for efficient and accurate Bayesian inversion using deep generative models. The method effectively tackles the curse of dimensionality and limited prior information, and produces accurate results with reliable uncertainty estimates.
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Remi Laumont, Valentin De Bortoli, Andres Almansa, Julie Delon, Alain Durmus, Marcelo Pereyra
Summary: This paper develops theory for Bayesian analysis and computation with PnP priors, introduces the PnP-ULA algorithm for Monte Carlo sampling and minimum mean square error estimation, establishes detailed convergence guarantees, and demonstrates that these algorithms approximately target a decision-theoretically optimal Bayesian model.
SIAM JOURNAL ON IMAGING SCIENCES
(2022)
Review
Multidisciplinary Sciences
Sylvia Fruehwirth-Schnatter
Summary: The paper discusses shrinkage priors that impose increasing shrinkage in a sequence of parameters. It introduces and extends the cumulative shrinkage process (CUSP) prior, which is a spike-and-slab shrinkage prior. The paper also demonstrates the representation of exchangeable spike-and-slab priors and introduces a new exchangeable spike-and-slab shrinkage prior based on the triple gamma prior.
PHILOSOPHICAL TRANSACTIONS OF THE ROYAL SOCIETY A-MATHEMATICAL PHYSICAL AND ENGINEERING SCIENCES
(2023)
Article
Physics, Multidisciplinary
Longxiang Liu, Lei Zhang, Xiaojun Tan, Youjin Deng
Summary: The study presents a family of graphical representations for the O(N) spin model and develops an efficient worm-type algorithm. It provides a convenient basis for studying spin models of different dimensions and can potentially be used in conjunction with advanced methods like tensor network renormalization.
COMMUNICATIONS IN THEORETICAL PHYSICS
(2023)
Article
Engineering, Mechanical
P. L. Green, L. J. Devlin, R. E. Moore, R. J. Jackson, J. Li, S. Maskell
Summary: This paper discusses the optimization of the 'L-kernel' in Sequential Monte Carlo samplers to improve performance, resulting in reduced variance of estimates and fewer resampling requirements.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Materials Science, Multidisciplinary
Wenxuan Zhang, Xiansong Xu, Zheyu Wu, Vinitha Balachandran, Dario Poletti
Summary: Neural network quantum states are powerful tools for analyzing complex quantum systems. We propose a local optimization procedure integrated with stochastic reconfiguration that outperforms global optimization approaches. We apply this method to analyze the ground state energy and correlations of the nonintegrable tilted Ising model with restricted Boltzmann machines. Sequential local updates lead to faster convergence to states closer to the ground state, depending on the size of the locally updated portion of the neural network. We demonstrate the generality of this approach by applying it to 1D and 2D nonintegrable spin systems.
Article
Computer Science, Interdisciplinary Applications
DanHua ShangGuan
Summary: The Monte Carlo method is a powerful tool in many research fields, but the increasing complexity of physical models and mathematical models requires efficient algorithms to overcome the computational cost.
JOURNAL OF COMPUTATIONAL PHYSICS
(2021)
Article
Mathematics, Interdisciplinary Applications
Yasuyuki Hamura, Kaoru Irie, Shonosuke Sugasawa
Summary: This paper discusses global-local shrinkage priors for analyzing count data, providing sufficient conditions under which the posterior mean is unshrunk, and proposing tractable priors to satisfy those conditions, as well as a custom posterior computation algorithm without tuning parameters.
Article
Engineering, Electrical & Electronic
Pierre Palud, Pierre-Antoine Thouvenin, Pierre Chainais, Emeric Bron, Franck Le Petit
Summary: This article focuses on a challenging class of inverse problems that arise in practical applications. The forward model is a complex non-linear black-box with potentially non-injective outputs spanning multiple decades. The observations are subject to both additive and multiplicative noises as well as censorship. The main objective of this work is to provide uncertainty quantification alongside parameter estimates using an adapted Bayesian approach and an MCMC algorithm to deal with the multimodal posterior distribution.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2023)
Article
Engineering, Mechanical
Adolphus Lye, Alice Cicirello, Edoardo Patelli
Summary: This tutorial paper reviews the use of advanced Monte Carlo sampling methods in Bayesian model updating for engineering applications, introducing different methods and comparing their performance. Three case studies demonstrate the advantages and limitations of these sampling techniques in parameter identification, posterior distribution sampling, and stochastic identification of model parameters.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Statistics & Probability
Yang Ni, Francesco C. Stingo, Veerabhadran Baladandayuthapani
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2019)
Article
Health Care Sciences & Services
Thierry Chekouo, Francesco C. Stingo, Caleb A. Class, Yuanqing Yan, Zachary Bohannan, Yue Wei, Guillermo Garcia-Manero, Samir Hanash, Kim-Anh Do
STATISTICAL METHODS IN MEDICAL RESEARCH
(2020)
Article
Biology
Christine B. Peterson, Nathan Osborne, Francesco C. Stingo, Pierrick Bourgeat, James D. Doecke, Marina Vannucci
Article
Statistics & Probability
Min Jin Ha, Francesco Claudio Stingo, Veerabhadran Baladandayuthapani
Summary: The integrative network modeling of data from multiple genomic platforms in this study sheds light on the multilayered genomic networks in human cancers. The Bayesian node-wise selection method outperforms existing methods in simulated data, offering potential for identifying biomarkers and therapeutic targets.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2021)
Article
Mathematical & Computational Biology
Federico Castelletti, Luca La Rocca, Stefano Peluso, Francesco C. Stingo, Guido Consonni
STATISTICS IN MEDICINE
(2020)
Article
Immunology
Elena Niccolai, Edda Russo, Simone Baldi, Federica Ricci, Giulia Nannini, Matteo Pedone, Francesco Claudio Stingo, Antonio Taddei, Maria Novella Ringressi, Paolo Bechi, Alessio Mengoni, Renato Fani, Giovanni Bacci, Camilla Fagorzi, Carolina Chiellini, Domenico Prisco, Matteo Ramazzotti, Amedeo Amedei
Summary: The study revealed higher levels of specific cytokines and bacteria genera in CRC tumor tissues compared to healthy mucosa, with a significant difference between the two groups. Additionally, the presence of certain bacterial genera was found to be positively or negatively correlated with the expression of IL-9 in the tumor microenvironment.
FRONTIERS IN IMMUNOLOGY
(2021)
Article
Nutrition & Dietetics
Simone Baldi, Marta Menicatti, Giulia Nannini, Elena Niccolai, Edda Russo, Federica Ricci, Marco Pallecchi, Francesca Romano, Matteo Pedone, Giovanni Poli, Daniela Renzi, Antonio Taddei, Antonino S. Calabro, Francesco C. Stingo, Gianluca Bartolucci, Amedeo Amedei
Summary: Changes in circulating levels of free fatty acids, including short chain, medium chain, and long chain fatty acids, are associated with metabolic, gastrointestinal, and malignant diseases. Analysis of FFA profiles in patients with celiac disease, adenomatous polyposis, and colorectal cancer compared to healthy controls revealed distinct differences, with butyric acid potentially serving as a biomarker for celiac disease screening. This study suggests gas chromatography-mass spectrometry as a suitable method for analyzing FFA composition in various gastrointestinal diseases.
Article
Oncology
Ichidai Tanaka, Delphine Dayde, Mei Chee Tai, Haruki Mori, Luisa M. Solis, Satyendra C. Tripathi, Johannes F. Fahrmann, Nese Unver, Gargy Parhy, Rekha Jain, Edwin R. Parra, Yoshiko Murakami, Clemente Aguilar-Bonavides, Barbara Mino, Muge Celiktas, Dilsher Dhillon, Julian Phillip Casabar, Masahiro Nakatochi, Francesco Stingo, Veera Baladandayuthapani, Hong Wang, Hiroyuki Katayama, Jennifer B. Dennison, Philip L. Lorenzi, Kim-Anh Do, Junya Fujimoto, Carmen Behrens, Edwin J. Ostrin, Jaime Rodriguez-Canales, Tetsunari Hase, Takayuki Fukui, Taisuke Kajino, Seiichi Kato, Yasushi Yatabe, Waki Hosoda, Koji Kawaguchi, Kohei Yokoi, Toyofumi F. Chen-Yoshikawa, Yoshinori Hasegawa, Adi F. Gazdar, Ignacio I. Wistuba, Samir Hanash, Ayumu Taguchi
Summary: SRGN is markedly overexpressed in TTF-1-negative LUAD cell lines, associated with poor clinical outcome, and influences PD-L1 expression and immune cell infiltration. SRGN expression is regulated by DNA demethylation and methionine metabolism.
JNCI-JOURNAL OF THE NATIONAL CANCER INSTITUTE
(2022)
Review
Statistics & Probability
Yang Ni, Veerabhadran Baladandayuthapani, Marina Vannucci, Francesco C. Stingo
Summary: Graphical models are powerful tools for investigating complex dependence structures in high-throughput biomedical datasets. Bayesian approaches are particularly suitable for large networks, as they encourage sparsity and uncertainty modeling. This paper reviews recent techniques for analyzing large networks under non-standard settings, demonstrating practical utility in cancer genomics and neuroimaging.
STATISTICAL METHODS AND APPLICATIONS
(2022)
Editorial Material
Statistics & Probability
Michael Schweinberger, Francesco C. Stingo, Maria Prosperina Vitale
Summary: The special issue showcases the breadth and depth of statistical learning with networks, covering observed and unobserved networks, and highlighting the usefulness of the network paradigm in addressing current problems through ten selected papers.
STATISTICAL METHODS AND APPLICATIONS
(2021)
Editorial Material
Statistics & Probability
Yang Ni, Veerabhadran Baladandayuthapani, Marina Vannucci, Francesco C. Stingo
STATISTICAL METHODS AND APPLICATIONS
(2022)
Article
Statistics & Probability
Matteo Pedone, Amedeo Amedei, Francesco C. Stingo
Summary: The composition of microbiota in different environments within the human body is linked to the development of various human diseases, including cancer. A recent study on colorectal cancer (CRC) motivated an investigation into the effect of clinical factors and diet-related covariates on the microbiota compositions. Through the development of a high-dimensional Bayesian hierarchical model, the proposed method identifies relevant associations and incorporates complex interactions. Simulation studies and analysis of CRC data demonstrate the benefits and superiority of the proposed approach over competing methods.
ANNALS OF APPLIED STATISTICS
(2023)
Article
Chemistry, Multidisciplinary
Gha Young Lee, Andrew A. Li, Intae Moon, Demos Katritsis, Yoannis Pantos, Francesco Stingo, Davide Fabbrico, Roberto Molinaro, Francesca Taraballi, Wei Tao, Claudia Corbo
Summary: Coronary artery disease (CAD) is a common and serious heart disease. Early detection is crucial and this study introduces a non-invasive diagnostic technology using nanotechnology to detect CAD. By forming disease-specific protein coronas on nanoparticles, a sensor array is developed for CAD detection. The results show high accuracy and sensitivity in detecting CAD using this method.
Article
Automation & Control Systems
Yang Ni, Francesco C. Stingo, Veerabhadran Baladandayuthapani
Summary: We introduce Bayesian Gaussian graphical models with covariates (GGMx), which are multivariate Gaussian distributions with covariate-dependent sparse precision matrix. We propose a general construction that maps the covariate space to sparse positive definite matrices, allowing for changes in strength and sparsity pattern of the precision matrix (graph structure) with the covariates. Our approach utilizes a novel mixture prior for precision matrices and ensures positive definiteness of sparse precision matrices using a carefully designed Markov chain Monte Carlo algorithm. Extensive simulations and a case study in cancer genomics demonstrate the utility of the proposed model.
JOURNAL OF MACHINE LEARNING RESEARCH
(2022)
Article
Mathematical & Computational Biology
Elin Shaddox, Christine B. Peterson, Francesco C. Stingo, Nicola A. Hanania, Charmion Cruickshank-Quinn, Katerina Kechris, Russell Bowler, Marina Vannucci