Partial least squares regression

Summary

Partial least squares regression (PLS regression) is a statistical method that bears some relation to principal components regression; instead of finding hyperplanes of maximum variance between the response and independent variables, it finds a linear regression model by projecting the predicted variables and the observable variables to a new space. Because both the X and Y data are projected to new spaces, the PLS family of methods are known as bilinear factor models. Partial least squares discriminant analysis (PLS-DA) is a variant used when the Y is categorical. PLS is used to find the fundamental relations between 2 matrices (X and Y), i.e. a latent variable approach to modeling the covariance structures in these two spaces. A PLS model will try to find the multidimensional direction in the X space that explains the maximum multidimensional variance direction in the Y space. PLS regression is particularly suited when the matrix of predictors has more variables than observations, a

Official source

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Related publications

Related people

Related units

Related concepts

Related courses

Related lectures

Remifentanil, like the other sedative drugs used to achieve conscious sedation, has an impact on the efficiency of patient ventilation. Since the quality of breathing has an indirect effect on blood O2 and CO2 concentrations, keeping patients' blood homeostasis relies on anesthetists' and nurses' vigilence. The current state of patient monitoring during conscious sedation can detect unsuccessful respirations only after they have occurred. Hence, there is a clinical need for a signal classifier to predict potential respiration failure in sedated patients. Given that the physiological changes involved in this context are transitory, and that opioids affect the Central Nervous System, the hypothesis is that opioids' effect on a patient drive to breath will result in detectable changes in electrocardiogram and capnogram signals. The purpose of this project is to design a signal classiffier to predict respiratory rate (RR) based on the analysis of the ECG spectrum and capnogram shape. Using patient audit data, a modeling set was created. Power Spectral Density calculation on a sliding Hamming window was the feature extraction used on ECG. A custom filter was designed to extract values describing the shape of each breaths in terms of slope, curvature and integral of the capnographic trace. Both patient specific (single user) and robust (mulit-user) solution were investigated. PLS regression resulted in parsimonious patient specific models capable of predicting the respiratory rate with an average accuracy of 90%. The performance, however, dropped for low RR. The analysis of the weights showed that the frequencies under 7.5Hz contain the majority of the information. Several limitations are addressed in the building of a robust multi-user model. The results obtained suggest an existing patient-specific correlation between the frequency content of ECG and the upcoming RR. Further investigation is required to achieve clinically useful prediction

2009

Bayesian Denoising: From MAP to MMSE Using Consistent Cycle Spinning

Emrah Bostan, Ulugbek Kamilov, Abbas Kazerouni, Michaël Unser

We introduce a new approach for the implementation of minimum mean-square error (MMSE) denoising for signals with decoupled derivatives. Our method casts the problem as a penalized least-squares regression in the redundant wavelet domain. It exploits the link between the discrete gradient and Haar-wavelet shrinkage with cycle spinning. The redundancy of the representation implies that some wavelet-domain estimates are inconsistent with the underlying signal model. However, by imposing additional constraints, our method finds wavelet-domain solutions that are mutually consistent. We confirm the MMSE performance of our method through statistical estimation of Levy processes that have sparse derivatives.

Ieee-Inst Electrical Electronics Engineers Inc2013

We consider the optimization of a quadratic objective function whose gradients are only accessible through a stochastic oracle that returns the gradient at any given point plus a zero-mean finite variance random error. We present the first algorithm that achieves jointly the optimal prediction error rates for least-squares regression, both in terms of forgetting the initial conditions in O (1 / n(2)), and in terms of dependence on the noise and dimension d of the problem, as O (d / n). Our new algorithm is based on averaged accelerated regularized gradient descent, and may also be analyzed through finer assumptions on initial conditions and the Hessian matrix, leading to dimension-free quantities that may still be small in some distances while the "optimal" terms above are large. In order to characterize the tightness of these new bounds, we consider an application to non-parametric regression and use the known lower bounds on the statistical performance (without computational limits), which happen to match our bounds obtained from a single pass on the data and thus show optimality of our algorithm in a wide variety of particular trade-offs between bias and variance.

2017