LOCO: Distributing Ridge Regression with Random Projections
نویسندگان
چکیده
We propose LOCO, a distributed algorithm which solves large-scale ridge regression. LOCO randomly assigns variables to different processing units which do not communicate. Important dependencies between variables are preserved using random projections which are cheap to compute. We show that LOCO has bounded approximation error compared to the exact ridge regression solution in the fixed design setting. Experimentally, in addition to obtaining significant speedups LOCO achieves the same predictive accuracy as standard ridge regression.
منابع مشابه
DUAL-LOCO: Distributing Statistical Estimation Using Random Projections
We present Dual-Loco, a communicatione cient algorithm for distributed statistical estimation. Dual-Loco assumes that the data is distributed across workers according to the features rather than the samples. It requires only a single round of communication where low-dimensional random projections are used to approximate the dependencies between features available to di↵erent workers. We show th...
متن کاملTwo-Parameters Fuzzy Ridge Regression with Crisp Input and Fuzzy Output
In this paper a new weighted fuzzy ridge regression method for a given set of crisp input and triangular fuzzy output values is proposed. In this regard, ridge estimator of fuzzy parameters is obtained for regression model and its prediction error is calculated by using the weighted fuzzy norm of crisp ridge coefficients. . To evaluate the proposed regression model, we introduce the fu...
متن کاملA MODIFICATION ON RIDGE ESTIMATION FOR FUZZY NONPARAMETRIC REGRESSION
This paper deals with ridge estimation of fuzzy nonparametric regression models using triangular fuzzy numbers. This estimation method is obtained by implementing ridge regression learning algorithm in the La- grangian dual space. The distance measure for fuzzy numbers that suggested by Diamond is used and the local linear smoothing technique with the cross- validation procedure for selecting t...
متن کاملa simulation comparison of Ridge regression estimators with Lars
This article has no abstract.
متن کاملNonlinear Forecasting with Many Predictors using Kernel Ridge Regression
This paper puts forward kernel ridge regression as an approach for forecasting with many predictors that are related nonlinearly to the target variable. In kernel ridge regression, the observed predictor variables are mapped nonlinearly into a high-dimensional space, where estimation of the predictive regression model is based on a shrinkage estimator to avoid overfitting. We extend the kernel ...
متن کامل