arXiv research

Researchers develop a method to infer reference measures from observed functionals.

This work extends entropic optimal transport to non-product reference couplings, focusing on Gaussian cases.

Using techniques of optimal transportation and gradient flows in metric spaces, we extend the notion of Riemannian Curvature Dimension condition $RCD(K,\infty)$ introduced (in case the reference measure is finite) by Giuseppe Savare', the first and the second author, to the case the reference measure is $σ$-finite; in …

This paper introduces a relative model risk measure of a product priced with a given model, with respect to another reference model for which the market is assumed to be driven. This measure allows comparing products valued with different models (pricing hypothesis) under a homogeneous framework which allows concluding…

Mondino and Naber recently proved that finite dimensional $\sf RCD$ spaces are rectifiable. Here we show that the push-forward of the reference measure under the charts built by them is absolutely continuous with respect to the Lebesgue measure. This result, read in conjunction with another recent work of us, has relev…

Study measures gender bias in machine translation using multiple reference points.

Decentralized learning achieves centralized performance via Gibbs measures.

We introduce a weak notion of barycenter of a probability measure $μ$ on a metric measure space $(X, d, {\bf m})$, with the metric $d$ and reference measure ${\bf m}$. Under the assumption that optimal transport plans are given by mappings, we prove that our barycenter $B(μ)$ is well defined; it is a probability measur…

Method identifies low-dimensional structure in high-dimensional probability measures.

Entropy asymmetry affects regularization in ERM, leading to biased solutions.

Existence proved for $q$-Bass martingales with specific marginals.

A novel approach to computing barycenters on graph-supported probability measures.

This work studies an explicit embedding of the set of probability measures into a Hilbert space, defined using optimal transport maps from a reference probability density. This embedding linearizes to some extent the 2-Wasserstein space, and enables the direct use of generic supervised and unsupervised learning algorit…

A framework for robust exploration in reinforcement learning under ambiguity.

In the presence of model risk, it is well-established to replace classical expected values by worst-case expectations over all models within a fixed radius from a given reference model. This is the "robustness" approach. We show that previous methods for measuring this radius, e.g. relative entropy or polynomial diverg…

Locally homogeneous RCD spaces are shown to be smooth manifolds.

Optimizes a portfolio for an investor preferring accepted securities over a reference security.

We study coherent risk measures which are time-consistent for multiple filtrations. We show that a coherent risk measure is time-consistent for every filtration if and only if it is one of four main types. Furthermore, if the risk measure is strictly monotone it is linear, and if the reference probability space is not …

New measures on orbit spaces for orthogonal groups identified.

Adjusted for chance measures are widely used to compare partitions/clusterings of the same data set. In particular, the Adjusted Rand Index (ARI) based on pair-counting, and the Adjusted Mutual Information (AMI) based on Shannon information theory are very popular in the clustering community. Nonetheless it is an open …

We show how risk measures originally defined in a model free framework in terms of acceptance sets and reference assets imply a meaningful underlying probability structure. Hereafter we construct a maximal domain of definition of the risk measure respecting the underlying ambiguity profile. We particularly emphasise li…

Study systemic risk measures adjusted to financial markets.

Stability of hypersurface immersions in Riemannian manifolds proved for $L^p$ perturbations.

Deep kernel learning refers to a Gaussian process that incorporates neural networks to improve the modelling of complex functions. We present a method that makes this approach feasible for problems where the data consists of line integral measurements of the target function. The performance is illustrated on computed t…

A new approach is presented to describe the change in the statistics of the log return distribution of financial data as a function of the timescale. To this purpose a measure is introduced, which quantifies the distance of a considered distribution to a reference distribution. The existence of a small timescale regime…

New regularization method reduces support of empirical risk minimization solutions.

Paper tackles measure estimation in barycentric coding model.

We show that gradient shrinking, expanding or steady Ricci solitons have potentials leading to suitable reference probability measures on the manifold. For shrinking solitons, as well as expanding soltions with nonnegative Ricci curvature, these reference measures satisfy sharp logarithmic Sobolev inequalities with low…

Model risk has a huge impact on any risk measurement procedure and its quantification is therefore a crucial step. In this paper, we introduce three quantitative measures of model risk when choosing a particular reference model within a given class: the absolute measure of model risk, the relative measure of model risk…

The study introduces measures of collective mobility from aggregated OD data.

The paper proposes a new method to measure risk with fine-grained tail sensitivity.

This article considers the quasi-local energy in reference to a general static spacetime. We follow the approach developed by the authors in [19, 20, 7, 9] and define the quasi-local energy as a difference of surface Hamiltonians, which are derived from the Einstein-Hilbert action. The new quasi-local energy provides a…

The paper introduces a new method to measure the shape relations between biological objects using r-parallel sets.

We investigate the low-dimensional structure of deterministic transformations between random variables, i.e., transport maps between probability measures. In the context of statistics and machine learning, these transformations can be used to couple a tractable "reference" measure (e.g., a standard Gaussian) with a tar…

Proves sufficiency of countable test plans for BV functions on metric spaces.

Machine learning employs dynamical algorithms that mimic the human capacity to learn, where the reinforcement learning ones are among the most similar to humans in this respect. On the other hand, adaptability is an essential aspect to perform any task efficiently in a changing environment, and it is fundamental for ma…

The paper provides precise estimates for isoperimetric inequalities on weighted manifolds.

Feedback alignment methods need to be evaluated for accuracy and gradient cosine similarity.

We introduce Gaussian-type measures on the manifold of all metrics with a fixed volume form on a compact Riemannian manifold of dimension $\geq 3$. For this random model we compute the characteristic function for the $L^2$ (Ebin) distance to the reference metric. In the Appendix, we study Lipschitz-type distance betwee…

This paper improves indoor positioning accuracy by deploying reference nodes to ensure Line-of-Sight.

One way to interpret smoothness of a measure in infinite dimensions is quasi-invariance of the measure under a class of transformations. Usually such settings lack a reference measure such as the Lebesgue or Haar measure, and therefore we can not use smoothness of a density with respect to such a measure. We describe h…

Active learning aims to train a classifier as fast as possible with as few labels as possible. The core element in virtually any active learning strategy is the criterion that measures the usefulness of the unlabeled data based on which new points to be labeled are picked. We propose a novel approach which we refer to …

We develop a tractable model of realization utility that studies the role of reference-dependent S-shaped preferences in a dynamic investment setting with reinvestment. Our model generates both voluntarily realized gains and losses. It makes specific predictions about the volume of gains and losses, the holding periods…

New algorithm improves signal reconstruction from noisy measurements with side information.

We refine and generalize several interpolation inequalities bounding the $L^p$ norm of a probability density with respect to the reference measure $μ$ by its Sobolev norm and the Kantorovich distance to $μ$ on a smooth weighted Riemannian manifold satisfying $CD(0, \infty)$ condition.

Bayesian approach to robust risk measures under model uncertainty.

Study dynamic risk measures with distributional uncertainty using optimal transport.

We show that compact fully connected (FC) deep learning networks trained to classify wireless protocols using a hierarchy of multiple denoising autoencoders (AEs) outperform reference FC networks trained in a typical way, i.e., with a stochastic gradient based optimization of a given FC architecture. Not only is the co…