For many real-world applications, we often need to select correlated variables---such as genetic variations and imaging features associated with Alzheimer's disease---in a high dimensional space. The correlation between variables presents a challenge to classical variable selection methods. To address this challenge, the elastic net has been developed and successfully applied to many applications. Despite its great success, the elastic net does not exploit the correlation information embedded in the data to select correlated variables. To overcome this limitation, we present a novel hybrid model, EigenNet, that uses the eigenstructures of data to guide variable selection.
Multiple Kernel Learning (MKL) generalizes SVMs to the setting where one simultaneously trains a linear classifier and chooses an optimal combination of given base kernels. Model complexity is typically controlled using various norm regularizations on the vector of base kernel mixing coefficients. Existing methods, however, neither regularize nor exploit potentially useful information pertaining to how kernels in the input set'interact'; that is, higher order kernel-pair relationships that can be easily obtained via unsupervised (similarity, geodesics), supervised (correlation in errors), or domain knowledge driven mechanisms (which features were used to construct the kernel?). We show that by substituting the norm penalty with an arbitrary quadratic function Q \succeq 0, one can impose a desired covariance structure on mixing coefficient selection, and use this as an inductive bias when learning the concept. This formulation significantly generalizes the widely used 1- and 2-norm MKL objectives.
Functional magnetic resonance imaging (fMRI) enables measuring human brain activity, in vivo. Yet, the fMRI hemodynamic response unfolds over very slow timescales ( 0.1-1 Hz), orders of magnitude slower than millisecond timescales of neural spiking. It is unclear, therefore, if slow dynamics as measured with fMRI are relevant for cognitive function. We investigated this question with a novel application of Gaussian Process Factor Analysis (GPFA) and machine learning to fMRI data. We analyzed slowly sampled (1.4 Hz) fMRI data from 1000 healthy human participants (Human Connectome Project database), and applied GPFA to reduce dimensionality and extract smooth latent dynamics.
The Continuous-Time Hidden Markov Model (CT-HMM) is an attractive approach to modeling disease progression due to its ability to describe noisy observations arriving irregularly in time. However, the lack of an efficient parameter learning algorithm for CT-HMM restricts its use to very small models or requires unrealistic constraints on the state transitions. In this paper, we present the first complete characterization of efficient EM-based learning methods for CT-HMM models. We demonstrate that the learning problem consists of two challenges: the estimation of posterior state probabilities and the computation of end-state conditioned statistics. We solve the first challenge by reformulating the estimation problem in terms of an equivalent discrete time-inhomogeneous hidden Markov model.
We propose a Bayesian mixed-effects model to learn typical scenarios of changes from longitudinal manifold-valued data, namely repeated measurements of the same objects or individuals at several points in time. The model allows to estimate a group-average trajectory in the space of measurements. Random variations of this trajectory result from spatiotemporal transformations, which allow changes in the direction of the trajectory and in the pace at which trajectories are followed. The use of the tools of Riemannian geometry allows to derive a generic algorithm for any kind of data with smooth constraints, which lie therefore on a Riemannian manifold. Stochastic approximations of the Expectation-Maximization algorithm is used to estimate the model parameters in this highly non-linear setting.The method is used to estimate a data-driven model of the progressive impairments of cognitive functions during the onset of Alzheimer's disease.
Linear regression models have been successfully used to function estimation and model selection in high-dimensional data analysis. However, most existing methods are built on least squares with the mean square error (MSE) criterion, which are sensitive to outliers and their performance may be degraded for heavy-tailed noise. In this paper, we go beyond this criterion by investigating the regularized modal regression from a statistical learning viewpoint. A new regularized modal regression model is proposed for estimation and variable selection, which is robust to outliers, heavy-tailed noise, and skewed noise. On the theoretical side, we establish the approximation estimate for learning the conditional mode function, the sparsity analysis for variable selection, and the robustness characterization.
A wide spectrum of discriminative methods is increasingly used in diverse applications for classification or regression tasks. However, many existing discriminative methods assume that the input data is nearly noise-free, which limits their applications to solve real-world problems. Particularly for disease diagnosis, the data acquired by the neuroimaging devices are always prone to different sources of noise. Robust discriminative models are somewhat scarce and only a few attempts have been made to make them robust against noise or outliers. These methods focus on detecting either the sample-outliers or feature-noises.
Evolving technology, best practice and landmark evidence in glaucoma care were reviewed by an international expert faculty in session presentations and debates during the 11th Moorfields International Glaucoma Symposium 2019. The authors were meeting chairs and provide an overview of symposium proceedings. Hans Lemij, Rotterdam Eye Hospital, the Netherlands, discussed glaucoma optical coherence tomography (OCT) imaging and automated segmentation issues, noting several common image artefacts. Paul Foster highlighted research by the UK Biobank Eye and Vision Consortium related to cognitive function and the expanding use of OCT imaging in dementia and neurodegeneration research. Findings show that a thinner retinal nerve fibre layer (RNFL) is associated with worse cognitive function in individuals without known neurodegenerative disease, as well as a greater likelihood of future cognitive decline . The Rotterdam Study also revealed an association of retinal neurodegeneration on OCT with an increased risk of dementia, including Alzheimer's disease .
Scientists have made artificial nerve cells, paving the way for new ways to repair the human body. The tiny "brain chips" behave like the real thing and could one day be used to treat diseases such as Alzheimer's. A team from the University of Bath used a combination of maths, computation and chip design to come up with a way to replicate in circuit form what nerve cells (neurons) do naturally. Neurons carry signals to and from the brain and the rest of the body. Scientists are interested in replicating them, because of the potential that offers in treating diseases such as Alzheimer's, where neurons degenerate or die.
It may be possible to delay your brain from aging. And no, this isn't the beginning of a pitch from an after-hours informercial -- the science behind this concept is surprisingly real. A recent study in Nature Neuroscience merged three fields to make strides in this research: longevity, neuroscience, and machine learning. An algorithm that can predict your brain age from MRI scans. Brain age refers to how well your brain is aging compared to your chronological age.