Sarbojit Roy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Sarbojit Roy has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Hernando Ombao
Statistics Program, King Abdullah University of Science and Technology, Thuwal 23955, Saudi Arabia
2 shared publications - Malik Shahid Sultan
Statistics Program, King Abdullah University of Science and Technology (KAUST), Thuwal 23955, Saudi Arabia
2 shared publications - Subhajit Dutta
Indian Institute of Technology Kanpur
2 shared publications - Anass B. El-Yaagoubi
King Abdullah University of Science and Technology
1 shared publication - Anil K. Ghosh
Theoretical Statistics and Mathematics Unit Indian Statistical Institute Kolkata India
1 shared publication - Aytijhya Saha
Indian Statistical Institute
1 shared publication
Latest Publications
- Open Access
Methods for Brain Connectivity Analysis with Applications to Rat Local Field Potential Recordings
Modeling the brain dependence network is central to understanding underlying neural mechanisms such as perception, action, and memory. In this study, we present a broad range of statistical methods for analyzing dependence in a brain network. Leveraging a combination of classical and cutting-edge approaches, we analyze multivariate hippocampal local field potential (LFP) time series data concentrating on the encoding of nonspatial olfactory information in rats. We present the strengths and limitations of each method in capturing neural dynamics and connectivity. Our analysis begins with exploratory techniques, including correlation, partial correlation, spectral matrices, and coherence, to establish foundational connectivity insights. We then investigate advanced methods such as Granger causality (GC), robust canonical coherence analysis, spectral transfer entropy (STE), and wavelet coherence to capture dynamic and nonlinear interactions. Additionally, we investigate the utility of topological data analysis (TDA) to extract multi-scale topological features and explore deep learning-based canonical correlation frameworks for connectivity modeling. This comprehensive approach offers an introduction to the state-of-the-art techniques for the analysis of dependence networks, emphasizing the unique strengths of various methodologies, addressing computational challenges, and paving the way for future research.
Open AccessMethods for Brain Connectivity Analysis with Applications to Rat Local Field Potential Recordings
Modeling the brain dependence network is central to understanding underlying neural mechanisms such as perception, action, and memory. In this study, we present a broad range of statistical methods for analyzing dependence in a brain network. Leveraging a combination of classical and cutting-edge approaches, we analyze multivariate hippocampal local field potential (LFP) time series data concentrating on the encoding of nonspatial olfactory information in rats. We present the strengths and limitations of each method in capturing neural dynamics and connectivity. Our analysis begins with exploratory techniques, including correlation, partial correlation, spectral matrices, and coherence, to establish foundational connectivity insights. We then investigate advanced methods such as Granger causality (GC), robust canonical coherence analysis, spectral transfer entropy (STE), and wavelet coherence to capture dynamic and nonlinear interactions. Additionally, we investigate the utility of topological data analysis (TDA) to extract multi-scale topological features and explore deep learning-based canonical correlation frameworks for connectivity modeling. This comprehensive approach offers an introduction to the state-of-the-art techniques for the analysis of dependence networks, emphasizing the unique strengths of various methodologies, addressing computational challenges, and paving the way for future research.
- Open Access
Classification of High-dimensional Time Series in Spectral Domain using Explainable Features
Interpretable classification of time series presents significant challenges in high dimensions. Traditional feature selection methods in the frequency domain often assume sparsity in spectral density matrices (SDMs) or their inverses, which can be restrictive for real-world applications. In this article, we propose a model-based approach for classifying high-dimensional stationary time series by assuming sparsity in the difference between inverse SDMs. Our approach emphasizes the interpretability of model parameters, making it especially suitable for fields like neuroscience, where understanding differences in brain network connectivity across various states is crucial. The estimators for model parameters demonstrate consistency under appropriate conditions. We further propose using standard deep learning optimizers for parameter estimation, employing techniques such as mini-batching and learning rate scheduling. Additionally, we introduce a method to screen the most discriminatory frequencies for classification, which exhibits the sure screening property under general conditions. The flexibility of the proposed model allows the significance of covariates to vary across frequencies, enabling nuanced inferences and deeper insights into the underlying problem. The novelty of our method lies in the interpretability of the model parameters, addressing critical needs in neuroscience. The proposed approaches have been evaluated on simulated examples and the `Alert-vs-Drowsy' EEG dataset.
Open AccessClassification of High-dimensional Time Series in Spectral Domain using Explainable Features
Interpretable classification of time series presents significant challenges in high dimensions. Traditional feature selection methods in the frequency domain often assume sparsity in spectral density matrices (SDMs) or their inverses, which can be restrictive for real-world applications. In this article, we propose a model-based approach for classifying high-dimensional stationary time series by assuming sparsity in the difference between inverse SDMs. Our approach emphasizes the interpretability of model parameters, making it especially suitable for fields like neuroscience, where understanding differences in brain network connectivity across various states is crucial. The estimators for model parameters demonstrate consistency under appropriate conditions. We further propose using standard deep learning optimizers for parameter estimation, employing techniques such as mini-batching and learning rate scheduling. Additionally, we introduce a method to screen the most discriminatory frequencies for classification, which exhibits the sure screening property under general conditions. The flexibility of the proposed model allows the significance of covariates to vary across frequencies, enabling nuanced inferences and deeper insights into the underlying problem. The novelty of our method lies in the interpretability of the model parameters, addressing critical needs in neuroscience. The proposed approaches have been evaluated on simulated examples and the `Alert-vs-Drowsy' EEG dataset.
- 27 November 2023
- Published by Taylor & Francis in Journal of Computational and Graphical Statistics
- Vol. 33 (2) , 448-462
On Exact Feature Screening in Ultrahigh-Dimensional Binary Classification
- 27 November 2023
- Published by Taylor & Francis in Journal of Computational and Graphical Statistics
- Vol. 33 (2) , 448-462
On Exact Feature Screening in Ultrahigh-Dimensional Binary Classification
- 18 September 2023
- Published by Springer Nature
- p. 86-101
Robust Classification of High-Dimensional Data Using Data-Adaptive Energy Distance
- 18 September 2023
- Published by Springer Nature
- p. 86-101
Robust Classification of High-Dimensional Data Using Data-Adaptive Energy Distance