Search papers, labs, and topics across Lattice.
This paper introduces Yau's Affine Normal Descent (YAND), a novel optimization framework leveraging the equi-affine normal of level-set hypersurfaces to define search directions invariant to volume-preserving affine transformations. The authors establish the equivalence of affine-normal directions with slice-centroid constructions for convex functions and demonstrate collinearity with Newton directions for strictly convex quadratics, leading to one-step convergence. They prove global, linear, and quadratic convergence under standard assumptions and validate the method's robustness to anisotropic scaling through numerical experiments.
Escape the tyranny of ill-conditioned optimization landscapes: Yau's Affine Normal Descent offers provably robust convergence by intrinsically adapting to anisotropic curvature through volume-preserving affine invariance.
We propose Yau's Affine Normal Descent (YAND), a geometric framework for smooth unconstrained optimization in which search directions are defined by the equi-affine normal of level-set hypersurfaces. The resulting directions are invariant under volume-preserving affine transformations and intrinsically adapt to anisotropic curvature. Using the analytic representation of the affine normal from affine differential geometry, we establish its equivalence with the classical slice-centroid construction under convexity. For strictly convex quadratic objectives, affine-normal directions are collinear with Newton directions, implying one-step convergence under exact line search. For general smooth (possibly nonconvex) objectives, we characterize precisely when affine-normal directions yield strict descent and develop a line-search-based YAND. We establish global convergence under standard smoothness assumptions, linear convergence under strong convexity and Polyak-Lojasiewicz conditions, and quadratic local convergence near nondegenerate minimizers. We further show that affine-normal directions are robust under affine scalings, remaining insensitive to arbitrarily ill-conditioned transformations. Numerical experiments illustrate the geometric behavior of the method and its robustness under strong anisotropic scaling.
