Trust-Region Algorithms for Training Responses: Machine Learning Methods Using Indefinite Hessian Approximations
This work addresses the challenge of hyper-parameter tuning and indefinite Hessian approximations in optimization for machine learning practitioners, though it appears incremental as it builds on existing quasi-Newton and trust-region frameworks.
The authors tackled the problem of training machine learning models by proposing a quasi-Newton trust-region method that allows indefinite Hessian approximations, achieving better results than limited-memory BFGS and Hessian-free methods on a standard dataset within a fixed computational time budget.
Machine learning (ML) problems are often posed as highly nonlinear and nonconvex unconstrained optimization problems. Methods for solving ML problems based on stochastic gradient descent are easily scaled for very large problems but may involve fine-tuning many hyper-parameters. Quasi-Newton approaches based on the limited-memory Broyden-Fletcher-Goldfarb-Shanno (BFGS) update typically do not require manually tuning hyper-parameters but suffer from approximating a potentially indefinite Hessian with a positive-definite matrix. Hessian-free methods leverage the ability to perform Hessian-vector multiplication without needing the entire Hessian matrix, but each iteration's complexity is significantly greater than quasi-Newton methods. In this paper we propose an alternative approach for solving ML problems based on a quasi-Newton trust-region framework for solving large-scale optimization problems that allow for indefinite Hessian approximations. Numerical experiments on a standard testing data set show that with a fixed computational time budget, the proposed methods achieve better results than the traditional limited-memory BFGS and the Hessian-free methods.