Online Statistical Inference for Parameters Estimation with Linear-Equality Constraints
This work addresses statistical inference for constrained optimization in machine learning, but it is incremental as it builds on existing PSGD methods.
The paper studied the limiting distribution of projected stochastic gradient descent (PSGD) estimates for parameters with linear-equality constraints, revealing how projection affects uncertainty, and proposed an online hypothesis testing procedure for these constraints, with confirmation through simulations and a real-world dataset.
Stochastic gradient descent (SGD) and projected stochastic gradient descent (PSGD) are scalable algorithms to compute model parameters in unconstrained and constrained optimization problems. In comparison with SGD, PSGD forces its iterative values into the constrained parameter space via projection. From a statistical point of view, this paper studies the limiting distribution of PSGD-based estimate when the true parameters satisfy some linear-equality constraints. Our theoretical findings reveal the role of projection played in the uncertainty of the PSGD-based estimate. As a byproduct, we propose an online hypothesis testing procedure to test the linear-equality constraints. Simulation studies on synthetic data and an application to a real-world dataset confirm our theory.