scholarly journals Doubly Sparsifying Network

2017 ◽  
Author(s):  
Zhangyang Wang ◽  
Shuai Huang ◽  
Jiayu Zhou ◽  
Thomas S. Huang
Keyword(s):  
Response Prediction ◽  
Low Complexity ◽  
Blood Oxygenation ◽  
Training Data ◽  
Superior Performance ◽  
Model Parameters ◽  
Unified Framework ◽  
Model Interpretation ◽  
Wide Range ◽  
Oxygenation Level

We propose the doubly sparsifying network (DSN), by drawing inspirations from the double sparsity model for dictionary learning. DSN emphasizes the joint utilization of both the problem structure and the parameter structure. It simultaneously sparsifies the output features and the learned model parameters, under one unified framework. DSN enjoys intuitive model interpretation, compact model size and low complexity. We compare DSN against a few carefully-designed baselines, to verify its consistently superior performance in a wide range of settings. Encouraged by its robustness to insufficient training data, we explore the applicability of DSN in brain signal processing that has been a challenging interdisciplinary area. DSN is evaluated for two mainstream tasks, electroencephalographic (EEG) signal classification and blood oxygenation level dependent (BOLD) response prediction, both achieving promising results.

A NEW MATHEMATICAL APPROACH FOR DETECTION OF ACTIVE AREA IN HUMAN BRAIN fMRI USING NONLINEAR MODEL

2010 ◽  
Vol 22 (05) ◽  
pp. 409-418 ◽  
Author(s):  
Ali Taalimi ◽  
Emad Fatemizadeh
Keyword(s):  
Moving Average ◽  
Simulated Data ◽  
Nonlinear Behavior ◽  
Blood Oxygenation ◽  
Superior Performance ◽  
Model Parameters ◽  
Bold Signal ◽  
Mathematical Approach ◽  
Region Detection ◽  
Oxygenation Level

Functional magnetic resonance imaging (fMRI) is widely-used for detection of the brain's neural activity. The signals and images acquired through this imaging technique demonstrate the human brain's response to pre-scheduled tasks. Several studies on blood oxygenation level-dependent (BOLD) signal responses demonstrate nonlinear behavior in response to a stimulus. In this paper we propose a new mathematical approach for modeling BOLD signal activity, which is able to model nonlinear and time variant behaviors of this physiological system. We employ the Nonlinear Auto Regressive Moving Average (NARMA) model to describe the mathematical relationship between output signals and predesigned tasks. The model parameters can be used to distinguish between rest and active states of a brain region. We applied our proposed method for active region detection on real as well as simulated data sets. The results show superior performance in comparison with existing methods.

scholarly journals Methods for Determining Frequency- and Region-Dependent Relationships Between Estimated LFPs and BOLD Responses in Humans

2009 ◽  
Vol 101 (1) ◽  
pp. 491-502 ◽  
Author(s):  
Roberto Martuzzi ◽  
Micah M. Murray ◽  
Reto A. Meuli ◽  
Jean-Philippe Thiran ◽  
Philippe P. Maeder ◽  
...  
Keyword(s):  
Human Subjects ◽  
Brain Regions ◽  
Source Model ◽  
Blood Oxygenation ◽  
Low Frequencies ◽  
Functional Regions ◽  
Wide Range ◽  
Bold Responses ◽  
Oxygenation Level ◽  
The Relationship

The relationship between electrophysiological and functional magnetic resonance imaging (fMRI) signals remains poorly understood. To date, studies have required invasive methods and have been limited to single functional regions and thus cannot account for possible variations across brain regions. Here we present a method that uses fMRI data and singe-trial electroencephalography (EEG) analyses to assess the spatial and spectral dependencies between the blood-oxygenation-level-dependent (BOLD) responses and the noninvasively estimated local field potentials (eLFPs) over a wide range of frequencies (0–256 Hz) throughout the entire brain volume. This method was applied in a study where human subjects completed separate fMRI and EEG sessions while performing a passive visual task. Intracranial LFPs were estimated from the scalp-recorded data using the ELECTRA source model. We compared statistical images from BOLD signals with statistical images of each frequency of the eLFPs. In agreement with previous studies in animals, we found a significant correspondence between LFP and BOLD statistical images in the gamma band (44–78 Hz) within primary visual cortices. In addition, significant correspondence was observed at low frequencies (<14 Hz) and also at very high frequencies (>100 Hz). Effects within extrastriate visual areas showed a different correspondence that not only included those frequency ranges observed in primary cortices but also additional frequencies. Results therefore suggest that the relationship between electrophysiological and hemodynamic signals thus might vary both as a function of frequency and anatomical region.

scholarly journals Vagal nerve stimulation triggers widespread responses and alters large-scale functional connectivity in the rat brain

2017 ◽  
Author(s):  
Jiayue Cao ◽  
Kun-Han Lu ◽  
Terry L. Powley ◽  
Zhongming Liu
Keyword(s):  
Vagus Nerve ◽  
Nerve Stimulation ◽  
Large Scale ◽  
Time Course ◽  
Brain Networks ◽  
Blood Oxygenation ◽  
7 Tesla ◽  
Sprague Dawley ◽  
Wide Range ◽  
Oxygenation Level

AbstractVagus nerve stimulation (VNS) is a therapy for epilepsy and depression. However, its efficacy varies and its mechanism remains unclear. Prior studies have used functional magnetic resonance imaging (fMRI) to map brain activations with VNS in human brains, but have reported inconsistent findings. The source of inconsistency is likely attributable to the complex temporal characteristics of VNS-evoked fMRI responses that cannot be fully explained by simplified response models in the conventional model-based analysis for activation mapping. To address this issue, we acquired 7-Tesla blood oxygenation level dependent fMRI data from anesthetized Sprague–Dawley rats receiving electrical stimulation at the left cervical vagus nerve. Using spatially independent component analysis, we identified 20 functional brain networks and detected the network-wise activations with VNS in a data-driven manner. Our results showed that VNS activated 15 out of 20 brain networks, and the activated regions covered >76% of the brain volume. The time course of the evoked response was complex and distinct across regions and networks. In addition, VNS altered the strengths and patterns of correlations among brain networks relative to those in the resting state. The most notable changes in network-network interactions were related to the limbic system. Together, such profound and widespread effects of VNS may underlie its unique potential for a wide range of therapeutics to relieve central or peripheral conditions.

scholarly journals Streaming stochastic variational Bayes: An improved approach for inference with concept drifting data streams

2019 ◽  
Author(s):  
Nadheesh Jihan ◽  
Malith Jayasinghe ◽  
Srinath Perera
Keyword(s):  
Online Learning ◽  
Data Streams ◽  
Bayesian Updating ◽  
Variational Inference ◽  
Variational Bayes ◽  
Superior Performance ◽  
Additional Analysis ◽  
Model Parameters ◽  
Posterior Variance ◽  
Wide Range

Online learning is an essential tool for predictive analysis based on continuous, endless data streams. Adopting Bayesian inference for online settings allows hierarchical modeling while representing the uncertainty of model parameters. Existing online inference techniques are motivated by either the traditional Bayesian updating or the stochastic optimizations. However, traditional Bayesian updating suffers from overconfident posteriors, where posterior variance becomes too inadequate to adapt to new changes to the posterior with concept drifting data streams. On the other hand, stochastic optimization of variational objective demands exhausting additional analysis to optimize a hyperparameter that controls the posterior variance. In this paper, we present "Streaming Stochastic Variational Bayes" (SSVB) — a novel online approximation inference framework for data streaming to address the aforementioned shortcomings of the current state-of-the-art. SSVB adjusts its posterior variance duly without any user-specified hyperparameters to control the posterior variance while efficiently accommodating the drifting patterns to the posteriors. Moreover, SSVB can be easily adopted by practitioners for a wide range of models (i.e. simple regression models to complex hierarchical models) with little additional analysis. We demonstrate the superior performance of SSVB against Population Variational Inference (PVI), Stochastic Variational Inference (SVI) and Black-box Streaming Variational Bayes (BB-SVB) using two non-conjugate probabilistic models: multinomial logistic regression and linear mixed effect model. Furthermore, we also emphasize the significant accuracy gain with SSVB based inference against conventional online learning models for each task.

scholarly journals Streaming stochastic variational Bayes: An improved approach for inference with concept drifting data streams

2019 ◽  
Author(s):  
Nadheesh Jihan ◽  
Malith Jayasinghe ◽  
Srinath Perera
Keyword(s):  
Online Learning ◽  
Data Streams ◽  
Bayesian Updating ◽  
Variational Inference ◽  
Variational Bayes ◽  
Superior Performance ◽  
Additional Analysis ◽  
Model Parameters ◽  
Posterior Variance ◽  
Wide Range

Online learning is an essential tool for predictive analysis based on continuous, endless data streams. Adopting Bayesian inference for online settings allows hierarchical modeling while representing the uncertainty of model parameters. Existing online inference techniques are motivated by either the traditional Bayesian updating or the stochastic optimizations. However, traditional Bayesian updating suffers from overconfident posteriors, where posterior variance becomes too inadequate to adapt to new changes to the posterior with concept drifting data streams. On the other hand, stochastic optimization of variational objective demands exhausting additional analysis to optimize a hyperparameter that controls the posterior variance. In this paper, we present "Streaming Stochastic Variational Bayes" (SSVB) — a novel online approximation inference framework for data streaming to address the aforementioned shortcomings of the current state-of-the-art. SSVB adjusts its posterior variance duly without any user-specified hyperparameters to control the posterior variance while efficiently accommodating the drifting patterns to the posteriors. Moreover, SSVB can be easily adopted by practitioners for a wide range of models (i.e. simple regression models to complex hierarchical models) with little additional analysis. We demonstrate the superior performance of SSVB against Population Variational Inference (PVI), Stochastic Variational Inference (SVI) and Black-box Streaming Variational Bayes (BB-SVB) using two non-conjugate probabilistic models: multinomial logistic regression and linear mixed effect model. Furthermore, we also emphasize the significant accuracy gain with SSVB based inference against conventional online learning models for each task.

scholarly journals DREAMTools: a Python package for scoring collaborative challenges

F1000Research ◽  
2015 ◽  
Vol 4 ◽  
pp. 1030 ◽  
Author(s):  
Thomas Cokelaer ◽  
Mukesh Bansal ◽  
Christopher Bare ◽  
Erhan Bilal ◽  
Brian M. Bot ◽  
...  
Keyword(s):  
Computational Methods ◽  
Training Data ◽  
Model Parameters ◽  
Automated Scoring ◽  
Statistical Machine Learning ◽  
Science And Engineering ◽  
New Methods ◽  
Wide Range ◽  
Improved Methods ◽  
Python Package

DREAM challenges are community competitions designed to advance computational methods and address fundamental questions in system biology and translational medicine. Each challenge asks participants to develop and apply computational methods to either predict unobserved outcomes or to identify unknown model parameters given a set of training data. Computational methods are evaluated using an automated scoring metric, scores are posted to a public leaderboard, and methods are published to facilitate community discussions on how to build improved methods. By engaging participants from a wide range of science and engineering backgrounds, DREAM challenges can comparatively evaluate a wide range of statistical, machine learning, and biophysical methods. Here, we describe DREAMTools, a Python package for evaluating DREAM challenge scoring metrics. DREAMTools provides a command line interface that enables researchers to test new methods on past challenges, as well as a framework for scoring new challenges. As of September 2015, DREAMTools includes more than 80% of completed DREAM challenges. DREAMTools complements the data, metadata, and software tools available at the DREAM website http://dreamchallenges.org and on the Synapse platform https://www.synapse.org.Availability: DREAMTools is a Python package. Releases and documentation are available at http://pypi.python.org/pypi/dreamtools. The source code is available at http://github.com/dreamtools.

scholarly journals DREAMTools: a Python package for scoring collaborative challenges

F1000Research ◽  
2016 ◽  
Vol 4 ◽  
pp. 1030 ◽  
Author(s):  
Thomas Cokelaer ◽  
Mukesh Bansal ◽  
Christopher Bare ◽  
Erhan Bilal ◽  
Brian M. Bot ◽  
...  
Keyword(s):  
Computational Methods ◽  
Training Data ◽  
Model Parameters ◽  
Automated Scoring ◽  
Statistical Machine Learning ◽  
Science And Engineering ◽  
Link Type ◽  
Wide Range ◽  
Improved Methods ◽  
Python Package

DREAM challenges are community competitions designed to advance computational methods and address fundamental questions in system biology and translational medicine. Each challenge asks participants to develop and apply computational methods to either predict unobserved outcomes or to identify unknown model parameters given a set of training data. Computational methods are evaluated using an automated scoring metric, scores are posted to a public leaderboard, and methods are published to facilitate community discussions on how to build improved methods. By engaging participants from a wide range of science and engineering backgrounds, DREAM challenges can comparatively evaluate a wide range of statistical, machine learning, and biophysical methods. Here, we describe DREAMTools, a Python package for evaluating DREAM challenge scoring metrics. DREAMTools provides a command line interface that enables researchers to test new methods on past challenges, as well as a framework for scoring new challenges. As of March 2016, DREAMTools includes more than 80% of completed DREAM challenges. DREAMTools complements the data, metadata, and software tools available at the DREAM website http://dreamchallenges.org and on the Synapse platform at https://www.synapse.org.Availability: DREAMTools is a Python package. Releases and documentation are available at http://pypi.python.org/pypi/dreamtools. The source code is available at http://github.com/dreamtools/dreamtools.

RESEARCH OF MULTI-TASK LEARNING BASED ON EXTREME LEARNING MACHINE

2013 ◽  
Vol 21 (supp02) ◽  
pp. 75-85 ◽  
Author(s):  
WENTAO MAO ◽  
JIUCHENG XU ◽  
SHENGJIE ZHAO ◽  
MEI TIAN
Keyword(s):  
Domain Knowledge ◽  
Real Life ◽  
Training Data ◽  
Learning Performance ◽  
Mixed Integer ◽  
Model Parameters ◽  
Minimization Method ◽  
Data Set ◽  
Task Learning ◽  
Wide Range

Recently, extreme learning machines (ELMs) have been a promising tool in solving a wide range of regression and classification applications. However, when modeling multiple related tasks in which only limited training data per task are available and the dimension is low, ELMs are generally hard to get impressive performance due to little help from the informative domain knowledge across tasks. To solve this problem, this paper extends ELM to the scenario of multi-task learning (MTL). First, based on the assumption that model parameters of related tasks are close to each other, a new regularization-based MTL algorithm for ELM is proposed to learn related tasks jointly via simple matrix inversion. For improving the learning performance, the algorithm proposed above is further formulated as a mixed integer programming in order to identify the grouping structure in which parameters are closer than others, and finally an alternating minimization method is presented to solve this optimization. Experiments conducted on a toy problem as well as real-life data set demonstrate the effectiveness of the proposed MTL algorithm compared to the classical ELM and the standard MTL algorithm.

scholarly journals Adversarial domain translation networks enable fast and accurate large-scale atlas-level single-cell data integration

2021 ◽  
Author(s):  
Jia Zhao ◽  
Gefei Wang ◽  
Jingsi Ming ◽  
Zhixiang Lin ◽  
Yang Wang ◽  
...  
Keyword(s):  
Single Cell ◽  
Rna Sequencing ◽  
Large Scale ◽  
Superior Performance ◽  
Brain Atlas ◽  
Unified Framework ◽  
Fine Grained ◽  
Wide Range ◽  
Shared Information ◽  
Single Nucleus

The rapid emergence of large-scale atlas-level single-cell RNA-sequencing (scRNA-seq) datasets from various sources presents remarkable opportunities for broad and deep biological investigations through integrative analyses. However, harmonizing such datasets requires integration approaches to be not only computationally scalable, but also capable of preserving a wide range of fine-grained cell populations. We created Portal, a unified framework of adversarial domain translation to learn harmonized representations of datasets. With innovation in model and algorithm designs, Portal achieves superior performance in preserving biological variation during integration, while having significantly reduced running time and memory compared to existing approaches, achieving integration of millions of cells in minutes with low memory consumption. We demonstrate the efficiency and accuracy of Portal using diverse datasets ranging from mouse brain atlas projects, the Tabula Muris project, and the Tabula Microcebus project. Portal has broad applicability and in addition to integrating multiple scRNA-seq datasets, it can also integrate scRNA-seq with single-nucleus RNA-sequencing (snRNA-seq) data. Finally, we demonstrate the utility of Portal by applying it to the integration of cross-species datasets with limited shared-information between them, and are able to elucidate biological insights into the similarities and divergences in the spermatogenesis process between mouse, macaque, and human.

scholarly journals Pattern Synthesis Using Multiple Kernel Learning for Efficient SVM Classification

2012 ◽  
Vol 12 (4) ◽  
pp. 77-94 ◽  
Author(s):  
Hari Seetha ◽  
R. Saravanan ◽  
M. Narasimha Murty
Keyword(s):  
Experimental Studies ◽  
Multiple Kernel Learning ◽  
Training Data ◽  
Superior Performance ◽  
Kernel Learning ◽  
Support Vector ◽  
Svm Classifier ◽  
Training Set ◽  
Multiple Kernel ◽  
Wide Range

Abstract Support Vector Machines (SVMs) have gained prominence because of their high generalization ability for a wide range of applications. However, the size of the training data that it requires to achieve a commendable performance becomes extremely large with increasing dimensionality using RBF and polynomial kernels. Synthesizing new training patterns curbs this effect. In this paper, we propose a novel multiple kernel learning approach to generate a synthetic training set which is larger than the original training set. This method is evaluated on seven of the benchmark datasets and experimental studies showed that SVM classifier trained with synthetic patterns has demonstrated superior performance over the traditional SVM classifier.

Sign in / Sign up

Export Citation Format

Share Document