Zeyi Xu

Mr. Zeyi Xu

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications

19 October 2025 in arXiv
Verified
Preprint
Open access
PDF file
LCLong Chen
Long Chen
ZXZeyi Xu
Zeyi Xu
We introduce and analyze two methods, HNAG+ and HNAG++, for minimizing strongly convex functions with large condition number kappa. For HNAG+, we prove a global linear convergence rate of 1 - 2/sqrt(kappa), achieving the information-theoretic optimal rate. For HNAG++, we establish a global asymptotic linear rate of 1 - 2*sqrt(2/kappa) for functions with Hölder continuous Hessians, representing the fastest known rate among globally convergent first-order methods. Extensive numerical experiments on linear and nonlinear problems show that HNAG++ consistently outperforms existing accelerated gradient methods.

27 January 2026 in arXiv
Verified
Preprint
Open access
PDF file
LCLong Chen
Long Chen
HLHao Luo
Hao Luo
JWJingrong Wei
Jingrong Wei
ZXZeyi Xu
Zeyi Xu
YYYuan Yao
Yuan Yao
Mirror descent uses the mirror function to encode geometry and constraints, improving convergence while preserving feasibility. Accelerated Mirror Descent Methods (Acc-MD) are derived from a discretization of an accelerated mirror ODE system using a variable--operator splitting framework. A geometric assumption, termed the Generalized Cauchy-Schwarz (GCS) condition, is introduced to quantify the compatibility between the objective and the mirror geometry, under which the first accelerated linear convergence for Acc-MD on a broad class of problems is established. Numerical experiments on smooth and composite optimization tasks demonstrate that Acc-MD consistently outperforms existing accelerated variants, both theoretically and empirically.

29 October 2026 in Journal of Optimization Theory and Applications
Verified
Journal Article
Open access
PDF file
LCLong Chen
Long Chen
ZXZeyi Xu
Zeyi Xu
Two accelerated first-order methods, $$\hbox {HNAG}^+$$ HNAG + and $$\hbox {HNAG}^{++}$$ HNAG + + , are introduced for smooth strongly convex optimization. They are derived from the Hessian-driven Nesterov Accelerated Gradient (HNAG) flow by optimizing the coercivity of shifted Lyapunov functions. Let $$\kappa =L/\mu $$ κ = L / μ , where $$\mu $$ μ is the strong-convexity constant and L is the gradient Lipschitz constant. To leading order as $$\kappa \rightarrow \infty $$ κ → ∞ , $$\hbox {HNAG}^+$$ HNAG + attains the global rate $$1-2/\sqrt{\kappa }$$ 1 - 2 / κ , matching the information-theoretic lower bound. Under local asymptotic symmetry at the minimizer, $$\hbox {HNAG}^{++}$$ HNAG + + achieves the faster asymptotic rate $$1-2\sqrt{2/\kappa }$$ 1 - 2 2 / κ to leading order. This matches the fastest known asymptotic rate for twice continuously differentiable objectives at leading order and extends it to a strictly larger function class. In particular, local asymptotic symmetry permits objectives that are not twice differentiable at the minimizer, showing that $${\mathcal {C}}^2$$ C 2 regularity is not necessary for this accelerated rate. A unified Lyapunov analysis establishes these guarantees, and numerical experiments confirm the predicted rates and show favorable performance against existing accelerated schemes.