{"id":"arxiv/2604.23922","url":"https://arxiv.org/abs/2604.23922","tags":["math.OC","cs.AI"],"score":0,"title":"Quasi-Quadratic Gradient: A New Direction for Accelerating the BFGS Method in Quasi-Newton Optimization","domain":"math.OC","source":"arxiv","authors":["John Chiang"],"reviews":0,"subject":"math.OC","summary":"In this paper, we introduce the Quasi-Quadratic Gradient (QQG), a novel search direction designed to accelerate the BFGS method within the quasi-Newton framework. By defining the QQG as the product of the inverse Hessian approximation and the current gradient, we explicitly leverage local second-order curvature to rectify the search path. Theoretical analysis and empirical results demonstrate that our approach significantly outperforms vanilla BFGS in convergence speed while maintaining computational efficiency.","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-04-27","reactions":{}}