Minimax Linear Regression under Measurement Constraints
نویسندگان
چکیده
We consider the problem of linear regression under measurement constraints and derive computationally feasible subsampling strategies to sample a small portion of design (data) points in a linear regression model y = Xβ + ε. The derived subsampling algorithms are minimax optimal for estimating the regression coefficients β under the fixed design setting, up to a small (1 + ) relative factor. Experiments on real-world data confirmed the effectiveness of our subsampling based linear regression algorithm with comparison to several other popular competitors. A longer technical report for this work can be found in (Wang & Singh, 2016).
منابع مشابه
Minimax Fixed-Design Linear Regression
We consider a linear regression game in which the covariates are known in advance: at each round, the learner predicts a real-value, the adversary reveals a label, and the learner incurs a squared error loss. The aim is to minimize the regret with respect to linear predictions. For a variety of constraints on the adversary’s labels, we show that the minimax optimal strategy is linear, with a pa...
متن کاملTruncated Linear Minimax Estimator of a Power of the Scale Parameter in a Lower- Bounded Parameter Space
Minimax estimation problems with restricted parameter space reached increasing interest within the last two decades Some authors derived minimax and admissible estimators of bounded parameters under squared error loss and scale invariant squared error loss In some truncated estimation problems the most natural estimator to be considered is the truncated version of a classic...
متن کاملMinimax Estimation of the Scale Parameter in a Family of Transformed Chi-Square Distributions under Asymmetric Squared Log Error and MLINEX Loss Functions
This paper is concerned with the problem of finding the minimax estimators of the scale parameter ? in a family of transformed chi-square distributions, under asymmetric squared log error (SLE) and modified linear exponential (MLINEX) loss functions, using the Lehmann Theorem [2]. Also we show that the results of Podder et al. [4] for Pareto distribution are a special case of our results for th...
متن کاملOptimal Quantum Measurement Design on Speech Signal: Blind Minimax Estimator Improving MSE Over LS Estimators
We consider the problem of estimating an unknown, deterministic speech signal parameters based on quantum measurements corrupted by white Gaussian noise. We design and analyze blind minimax estimator (BME), which consist of a bounded parameter set. Using minimax estimator, the parameter set is itself estimated from quantum measurements. Thus, our approach does not require any prior knowledge of...
متن کاملRate Minimaxity of the Lasso and Dantzig Estimators
We consider the estimation of regression coefficients in a high-dimensional linear model. A lower bound of the minimax `q risk is provided for regression coefficients in `r balls, along with a minimax lower bound for the tail of the `q loss. Under certain conditions on the design matrix and penalty level, we prove that these minimax convergence rates are attained by both the Lasso and Dantzig e...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2016