arXiv research

Introduces bounded scale measure and generalizes property A.

We study generalizations of Reifenberg's Theorem for measures in $\mathbb R^n$ under assumptions on the Jones' $β$-numbers, which appropriately measure how close the support is to being contained in a subspace. Our main results, which holds for general measures without density assumptions, give effective measure bounds…

New algorithms reduce contextual bandits' regret without knowing reward noise variances.

The study shows how to measure translation surfaces with short saddle connections.

We show that if $X$ is a limit of $n$-dimensional Riemannian manifolds with Ricci curvature bounded below and $γ$ is a limit geodesic in $X$ then along the interior of $γ$ same scale measure metric tangent cones $T_{γ(t)}X$ are Hölder continuous with respect to measured Gromov-Hausdorff topology and have the same dimen…

For collapsing sequences of Riemannian manifolds which satisfy a uniform lower Ricci curvature bound it is shown that there is a sequence of scales such that for a set of good base points of large measure the pointed rescaled manifolds subconverge to a product of a Euclidean and a compact space. All Euclidean factors h…

Existing strategies for finite-armed stochastic bandits mostly depend on a parameter of scale that must be known in advance. Sometimes this is in the form of a bound on the payoffs, or the knowledge of a variance or subgaussian parameter. The notable exceptions are the analysis of Gaussian bandits with unknown mean and…

Despite existing work on ensuring generalization of neural networks in terms of scale sensitive complexity measures, such as norms, margin and sharpness, these complexity measures do not offer an explanation of why neural networks generalize better with over-parametrization. In this work we suggest a novel complexity m…

New bounds on self-normalized martingales improve online linear regression performance.

It has recently been shown that for compressive sensing, significantly fewer measurements may be required if the sparsity assumption is replaced by the assumption the unknown vector lies near the range of a suitably-chosen generative model. In particular, in (Bora {\em et al.}, 2017) it was shown roughly $O(k\log L)$ r…

Improved graph neural network bounds using graph diffusion matrix.

ASAM improves deep neural network generalization by adapting sharpness to scale.

Study dynamic risk measures with distributional uncertainty using optimal transport.

This research evaluates generalization measures in deep learning.

The paper provides bounds for the empirical angular measure and applies them to improve statistical learning in extreme regions.

In this note, we prove a uniform distance distortion estimate for Ricci flows with uniformly bounded scalar curvature, independent of the lower bound of the initial $μ$-entropy. Our basic principle tells that once correctly renormalized, the metric-measure quantities obey similar estimates as in the non-collapsing case…

Develops a new method for neural network significance testing without strict constraints.

This paper proves a generalization bound for complex-valued neural networks scaling with spectral complexity.

Generalization of deep networks has been of great interest in recent years, resulting in a number of theoretically and empirically motivated complexity measures. However, most papers proposing such measures study only a small set of models, leaving open the question of whether the conclusion drawn from those experiment…

Upper bound found for systolic geometry on manifolds with positive scalar curvature.

Unified CMI bounds for meta learning, bridging information-theoretic and classical learning theory.

Paper analyzes convergence of two time-scale stochastic approximation using martingale approach.

Measure-scaling quasi-isometries on graphs have specific scaling groups.

Paper investigates separating times for general diffusions, providing new insights.

This paper reviews the economic and theoretical foundations of insolvency risk measurement and capital adequacy rules. The proposed new measure of insolvency risk is constructed by disentangling assets, debt and equity at the micro-prudential firm level. This new risk index is the Firm Insolvency Risk Index (FIRI) whic…

New method removes scalar curvature assumption in Ricci flow smoothing.

In 1996, Shi generalized the epsilon-regularity theorem of Schoen and Uhlenbeck to energy-minimizing harmonic maps from a domain equipped with a bounded measurable Riemannian metric. In the present work we prove a compactness result for such energy-minimizing maps. As an application, we combine our result with Shi's th…

New summary measures reveal geometric structure in weighted measures on manifolds.

The paper identifies a 'small' set of functions containing Gaussian process samples.

New Fourier features improve high-precision approximation in large-scale problems.

We give sufficient conditions for a measured length space (X,d,m) to admit local and global Poincare inequalities. We first introduce a condition DM on (X,d,m), defined in terms of transport of measures. We show that DM, along with a doubling condition on m, implies a scale-invariant local Poincare inequality. We show …

Generalized algorithm for translation and scale-invariant prediction.

Uniform scaling limits in AdamW-trained transformers converge to ODEs.

Paper proposes data quality measures for large-scale high-dimensional data.

Study risk-sensitive reinforcement learning with entropic risk measures and generative models.

A new measure predicts deep learning model performance.

Minimizing a convex function of a measure with a sparsity-inducing penalty is a typical problem arising, e.g., in sparse spikes deconvolution or two-layer neural networks training. We show that this problem can be solved by discretizing the measure and running non-convex gradient descent on the positions and weights of…

These series of notes serve as an introduction to some of both the classical and modern techniques in Reifenberg theory. At its heart, Reifenberg theory is about studying general sets or measures which can be, in one sense or another, approximated on all scales by well behaved spaces, typically just Euclidean space its…

New algorithm bounds regret in mediator feedback bandit problems.

The paper sets criteria for no arbitrage in complex financial models.

Kernel interpolation is inconsistent for norms with smoothness above a constant.

New algorithm handles bandit problems under translations and scales.

We define a new family of similarity and distance measures on graphs, and explore their theoretical properties in comparison to conventional distance metrics. These measures are defined by the solution(s) to an optimization problem which attempts find a map minimizing the discrepancy between two graph Laplacian exponen…

In safety-critical applications a probabilistic model is usually required to be calibrated, i.e., to capture the uncertainty of its predictions accurately. In multi-class classification, calibration of the most confident predictions only is often not sufficient. We propose and study calibration measures for multi-class…

Introduces new performance measures using scaled utility functions.

Kernel dependence measures yield accurate estimates of nonlinear relations between random variables, and they are also endorsed with solid theoretical properties and convergence rates. Besides, the empirical estimates are easy to compute in closed form just involving linear algebra operations. However, they are hampere…

We consider Markov Decision Processes (MDPs) where the rewards are unknown and may change in an adversarial manner. We provide an algorithm that achieves state-of-the-art regret bound of $O( \sqrt{τ(\ln|S|+\ln|A|)T}\ln(T))$, where $S$ is the state space, $A$ is the action space, $τ$ is the mixing time of the MDP, and $…

Quantum-enhanced metrology aims to estimate an unknown parameter such that the precision scales better than the shot-noise bound. Single-shot adaptive quantum-enhanced metrology (AQEM) is a promising approach that uses feedback to tweak the quantum process according to previous measurement outcomes. Techniques and form…