Balancing Statistical and Computational Precision: A General Theory and Applications to Sparse Regression
This addresses the challenge of handling large datasets efficiently for researchers and practitioners in machine learning, though it appears incremental as it builds on existing regularized estimation methods.
The paper tackles the problem of balancing statistical soundness and computational efficiency in regularized estimation, showing that their approach improves on standard pipelines for sparse and group-sparse regression.
Modern technologies are generating ever-increasing amounts of data. Making use of these data requires methods that are both statistically sound and computationally efficient. Typically, the statistical and computational aspects are treated separately. In this paper, we propose an approach to entangle these two aspects in the context of regularized estimation. Applying our approach to sparse and group-sparse regression, we show that it can improve on standard pipelines both statistically and computationally.