Stability of submanifold cut loci under metric perturbations proved.
problem Stability of submanifold cut loci under metric perturbations.
method Continuity of injectivity radius and Whitney C2 perturbation of submanifolds. result Hausdorff stability of submanifold cut loci under C2 metric perturbations. For a polygonal knot K, it is shown that a tube of radius R(K), the polygonal thickness radius, is an embedded torus. Given a thick configuration K, perturbations of size r<R(K) define satellite structures, or local knotting. We explore knotting within these tubes both theoretically and numerically. We provide bounds o…
Logarithmic network width suffices for robust memorization.
problem Achieving robust memorization in neural networks.
method Established upper and lower bounds on robust memorization radius.
result Width logarithmic in the number of samples is necessary and sufficient for robust memorization.
New uniqueness concept for adversarial Bayes classifier.
problem Understanding adversarial Bayes classifiers in binary classification.
method Developed a new notion of uniqueness and analyzed it for a family of one-dimensional data distributions.
result Improved regularity of adversarial Bayes classifiers as perturbation radius increases.
In this paper, we discuss the sensitivity of quantum PageRank. By using the finite dimensional perturbation theory, we estimate the change of the quantum PageRank under a small analytical perturbation on the Google matrix. In addition, we will show the way to estimate the lower bound of the convergence radius as well a…
Adversarial training purifies hidden weights to remove small perturbations.
problem Understanding and removing adversarial perturbations in deep learning models.
method Introducing Feature Purification, a principle that adversarial training aims to remove small dense mixtures in hidden weights.
result Adversarial training can make neural networks robust against small perturbations, even with simple algorithms.
Adversarial training improves linear regression solutions, offering robustness against small perturbations.
problem Vulnerability of linear models to adversarial perturbations.
method Formulated as a min-max problem, adversarial training minimizes the best solution under worst-case attacks.
result Adversarial training yields the minimum-norm interpolating solution in overparameterized models, equivalent to parameter shrinking methods in underparameterized models.
New method learns robustly with less data, bridging theory and practice.
problem Adversarial robust learning with metric perturbation.
method Tolerant adversarial PAC-learning with perturb-and-smooth approach and compression-based algorithm.
result First PAC-type guarantees for popular adversarial learning techniques.
Estimates mass of static vacuum metrics with small Bartnik data.
problem Estimating mass of static vacuum metrics with small perturbations.
method Second-order mass estimation using Bartnik data.
result New upper bound on Bartnik mass to fifth order.
On a smooth complete Riemannian spin manifold with smooth compact boundary, we demonstrate that the Atiyah-Singer Dirac operator DB in L2 depends Riesz continuously on L∞ perturbations of local boundary conditions B. The Lipschitz bound for the map ${…
We prove that the Atiyah-Singer Dirac operator Dg in L2 depends Riesz continuously on L∞ perturbations of complete metrics g on a smooth manifold. The Lipschitz bound for the map ${\mathrm g} \to {\mathrm D}_{\mathrm g}(1 + {\mathrm D}_{\mathrm g}^2)^{…
In this paper we study the behaviour of the continuous spectrum of the Laplacian on a complete Riemannian manifold of bounded curvature under perturbations of the metric. The perturbations that we consider are such that its covariant derivatives up to some order decay with some rate in the geodesic distance from a fixe…
New method μP2 improves neural network training by scaling perturbations layerwise.
problem Improving neural network performance as models scale up.
method Layerwise perturbation scaling in the infinite-width limit of neural networks.
result Layerwise perturbation scaling ensures all layers are effectively perturbed in the limit.
Improved neural network robustness with instance-specific perturbation margins.
problem Adversarial training fails to generalize well to unperturbed test set.
method Instance adaptive adversarial training with sample-specific perturbation margins.
result Test accuracy improves with a marginal drop in robustness.
We introduce RSE to measure robustness in estimation problems.
problem Estimating statistical models from observed data.
method Developed theory for spectral functions of measures to compute RSE.
result RSE reveals a reciprocal relationship with problem complexity.
The gluing technique is used to construct hypersurfaces in Euclidean space having approximately constant prescribed mean curvature. These surfaces are perturbations of unions of finitely many spheres of the same radius assembled end-to-end along a line segment. The condition on the existence of these hypersurfaces is t…
It was recently shown by R. Souam and E. Toubiana that the (non constantly curved) Berger spheres do not contain totally umbilic surfaces. Nevertheless in this article we show, by perturbative arguments, that all analytic metrics sufficiently close to the round metric g0 on S3 possess \textsl{general…
Despite the improved accuracy of deep neural networks, the discovery of adversarial examples has raised serious safety concerns. In this paper, we study two variants of pointwise robustness, the maximum safe radius problem, which for a given input sample computes the minimum distance to an adversarial example, and the …
Owing to the susceptibility of deep learning systems to adversarial attacks, there has been a great deal of work in developing (both empirically and certifiably) robust classifiers. While most work has defended against a single type of attack, recent work has looked at defending against multiple perturbation models usi…
In this work, we investigate black-box optimization from the perspective of frequentist kernel methods. We propose a novel batch optimization algorithm, which jointly maximizes the acquisition function and select points from a whole batch in a holistic way. Theoretically, we derive regret bounds for both the noise-free…
Extends randomized smoothing to certify robustness against various threat models and adversarial perturbations.
problem Certifying robustness of classifiers against adversarial perturbations.
method Develops a method to certify robustness against any ℓp (p∈N>0) minimized adversarial perturbation. result Randomized smoothing suffers from the curse of dimensionality, reducing effective radius as p increases. Study how nodal domains change on surfaces under perturbations.
problem How eigenfunction nodal domains change on surfaces under smooth perturbations.
method Sector/graph count near nodal critical points, upper semicontinuity proof, branch-free on spectral clusters, wavelength-scale analysis.
result Upper semicontinuity of nodal domain count, no new domains created at wavelength scale, stable count in noncritical cases.
We show, using a theorem of Milnor and Margulis, that string theory on compact negatively curved spaces grows new effective dimensions as the space shrinks, generalizing and contextualizing the results in hep-th/0510044. Milnor's theorem relates negative sectional curvature on a compact Riemannian manifold to exponenti…
Given (M, g0) we consider the problem -ε^2Delta_{g0+h}u + u = (u+)^{p-1} with (ε, h) \in (0, ε0) \times Bρ. Here Bρ is a ball centered at 0 with radius ρ in the Banach space of all Ck symmetric covariant 2-tensors on M. Using the Poincaré polynomial of M, we give an estimate on the number of nonconstant solutions with …
Let (M,g) be a compact Riemannian manifold of dimension N≥2. We prove the existence of a family (Ωε)ε∈(0,ε0) of self-Cheeger sets in (M,g) . The domains Ωε⊂M are perturbations of geodesic balls of radius ε c…
Let (M,g) be a compact Riemannian manifold of dimension N≥2. We prove the existence of a family (Ωε)ε∈(0,ε0) of self-Cheeger sets in (M,g) . The domains Ωε⊂M are perturbations of geodesic balls of radius ε cen…
Label noise in adversarial training leads to robust overfitting, explained and mitigated.
problem Label noise in adversarial training causes robust overfitting.
method Proposed a method to automatically calibrate labels.
result Consistent performance improvements across various models and datasets.
We prove a general theorem providing smoothed analysis estimates for conic condition numbers of problems of numerical analysis. Our probability estimates depend only on geometric invariants of the corresponding sets of ill-posed inputs. Several applications to linear and polynomial equation solving show that the estima…
In this paper we prove the following pointwise and curvature-free estimates on convexity radius, injectivity radius and local behavior of geodesics in a complete Riemannian manifold M: 1) the convexity radius of p, $\operatorname{conv}(p)\ge \min\{\frac{1}{2}\operatorname{inj}(p),\operatorname{foc}(B_{\operatorname…
Positive injectivity radius for manifolds with Lie structure at infinity.
problem Injectivity radius positivity for manifolds with specific boundary conditions.
method Lie groupoids to prove injectivity radius positivity.
result Injectivity radius is positive for manifolds with Lie structure at infinity.
The ratio of convexity radius over injectivity radius may be made arbitrarily small within the class of compact Riemannian manifolds of any fixed dimension at least two. This is proved using Gulliver's method of constructing manifolds with focal points but no conjugate points. The approach is suggested by a characteriz…
This paper is concerned with the location of nodal sets of eigenfunctions of the Dirichlet Laplacian in thin tubular neighbourhoods of hypersurfaces of the Euclidean space of arbitrary dimension. In the limit when the radius of the neighbourhood tends to zero, it is known that spectral properties of the Laplacian are a…
Uniform curvature bounds for regularized metrics with bounds on Ricci tensor and injectivity radius.
problem Bounding curvature of regularized metrics with constraints on Ricci tensor and injectivity radius.
method Mollification of riemannian metrics, uniform W2,p-harmonic radius bounds, Ricci tensor bounds, injectivity radius bounds. result Uniform estimate on the change of sectional curvature for regularized metrics.
Upper bound for conjugate radius in open manifolds with scalar curvature and spectrum constraints.
problem Bounding the conjugate radius of open manifolds with specific curvature and spectrum conditions.
method Established an upper bound using scalar curvature and bottom-of-spectrum constraints.
result For certain conditions, the conjugate radius is no more than π.
Proves upper bound on filling radius for manifolds with positive scalar curvature.
problem Bounding the filling radius of manifolds with positive scalar curvature.
method Quantitative operator K-theory and index theory.
result Proves a quantitative upper bound on the filling radius.
Adversarial training leads to large generalization gap, decomposed into bias and variance.
problem Understanding the large generalization gap in adversarially trained models.
method Bias-Variance decomposition of test risk as a function of adversarial perturbation radius.
result Bias increases monotonically with adversarial perturbation radius and is dominant in test risk.
Compact theorem for minimal surfaces with lower injectivity radius.
problem Proving compactness of minimal surfaces with lower injectivity radius.
method Variant of Choi--Schoen compactness theorem, focusing on injectivity radius.
result Proved compactness theorem for minimal surfaces.
Our attacks are stronger and faster under Wasserstein metric.
problem Vulnerability of deep models to adversarial attacks.
method Developed an exact yet efficient projection operator and used the Frank-Wolfe method.
result Generated much stronger attacks and improved model robustness.
Injectivity radius on Stiefel manifold is π.
problem Determining the injectivity radius of the compact Stiefel manifold.
method Utilized the property of geodesics being space curves of constant Frenet curvatures.
result The injectivity radius on the Stiefel manifold under the Euclidean metric is π.
Lower bound on boundary injectivity radius for specific tubes.
problem Estimating the boundary injectivity radius of Margulis tubes.
method Using curvature bounds to derive a lower bound.
result A lower bound on the boundary injectivity radius is provided.
Study gives bounds on filling radius for Riemannian manifolds.
problem Finding bounds on the filling radius of Riemannian manifolds.
method Curvature-dependent bounds for all closed manifolds and upper bounds for submersion and submetry cases.
result Upper and lower bounds on the filling radius for specific types of manifolds.
Validates neural networks inputs to protect against adversarial examples.
problem Ensuring neural networks robustness against adversarial attacks.
method Runtime local robustness verification based on normal distribution of robustness radii.
result Improves neural network accuracy and protects against adversarial examples.
The paper proves estimates and theorems for Kähler manifolds.
problem Curvature conditions on Kähler manifolds.
method Volume comparison and rigidity theorems.
result Conjugate radius estimates for Kähler manifolds.
Study finds the covering radius of RM(4,8) is 26.
problem Determining the covering radius of RM(4,8).
method Invented a lift by derivation invariant to classify B(5,6,8).
result Covering radius of RM(4,8) is 26.
Clarifies definition of polarized canonical radius in Kahler Ricci flow.
problem Unclear definition of polarized canonical radius in Kahler Ricci flow.
method Clarification of the definition.
result Clarified definition of polarized canonical radius.
The paper bounds bandwidth and focal radius for manifolds with positive isotropic curvature.
problem Bounding bandwidth and focal radius for manifolds with positive isotropic curvature.
method Using spectral properties of a twisted de Rham-Hodge operator.
result Upper bounds on bandwidth and focal radius are derived for hypersurfaces in PIC manifolds.
Upper bound on Stiefel manifold's injectivity radius found.
problem Finding the maximum distance within which the Stiefel manifold remains injective.
method Exhibited conjugate points and calculated the minimum of geodesic lengths.
result Upper bound on Stiefel manifold's injectivity radius is conjectured to be equal to the injectivity radius.
Upper estimate of Dirac eigenvalue linked to hyperspherical radius.
problem Estimating the smallest eigenvalue of the Dirac operator.
method Proved an upper estimate of the smallest eigenvalue in terms of hyperspherical radius.
result Combining with known lower estimates, geometric consequences are derived.