A Stochastic-Gradient-based Interior-Point Algorithm for Solving Smooth Bound-Constrained Optimization Problems

arXiv:2304.14907v2 Announce Type: replace-cross
Abstract: A stochastic-gradient-based interior-point algorithm for minimizing a continuously differentiable objective function (that may be nonconvex) subject to bound constraints is presented, analyzed, and demonstrated through experimental results. The algorithm is unique from other interior-point methods for solving smooth \edit{nonconvex} optimization problems since the search directions are computed using stochastic gradient estimates. It is also unique in its use of inner neighborhoods of the feasible region -- defined by a positive and vanishing neighborhood-parameter sequence -- …

abstract algorithm arxiv constraints cs.lg differentiable edit experimental function gradient math.oc optimization results search stochastic stochastic-gradient the algorithm through type