An Improved SQP Method for Solving Unconstrained Minimax Problem
Lei Shi
Abstract
Lei Shi
Abstract
The unconstrained minimax problems are discussed in this paper. By using normrelaxed sequential quadratic programming( SQP) method and combining the active set identification technique,a perturbed SQP method without hypothesis of positive definite matrix is proposed. Under mild conditions,the algorithm has global convergence. Preliminary numerical results show that the algorithm is effective.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The unconstrained minimax problems are discussed in this paper. By using normrelaxed sequential quadratic programming( SQP) method and combining the active set identification technique,a perturbed SQP method without hypothesis of positive definite matrix is proposed. Under mild conditions,the algorithm has global convergence. Preliminary numerical results show that the algorithm is effective.
Key concepts: Sequential quadratic programming, Minimax, Mathematical optimization, Convergence (economics), Mathematics, Quadratic programming, Set (abstract data type), Computer science