Searcharxiv⌕ Search

arXiv subjects

Oliver Stein

Publications and source records attributed to Oliver Stein.

11 recordsLinked to original sources
arXiv2026

A single loop method for quadratic minmax optimization

We consider a quadratic minmax problem with coupled inner constraints and propose a method to compute a class of stationary points. To motivate the need to compute such stationary points, we first show that they are meaningful, in the sense that they can be locally optimal for our problem under suitable{non-degeneracy} conditions. Then based on a suitable log barrier function, we build an infeasible interior point-type {single loop method} (which does not explicitly distinguish between the outer and inner problem) and prove that a non-degenerate stationary point is an attraction point as the algorithm moves along the designed central path. We show in particular that our method is polynomial in the special case where the inner feasible set of our constrained minmax problem is independent from outer variables. Our numerical experiments, on both synthetic data and a class of min-cost flow problems, showcase the behavior of our method and how it outperforms existing algorithms from the literature in terms of the quality of the computed stationary points.

math.OC↗
arXiv2024

The standard $L$-function attached to a vector valued modular form

We define two $L$-functions associated to a common vector valued eigenform $f$ transforming with the ``finite'' Weil representation. The first one can be seen as a standard zeta function defined by the eigenvalues of $f$. The second one can be interpreted as standard $L$-function defined as an Euler product where each $p$-factor is a rational function in terms of two unramified characters of the $p$-adic field $\Q_p$. We show that both $L$-functions are related and prove further that they both can be continued meromorphically to the whole complex $s$-plane.

math.NT↗
arXiv2023

A converse theorem for Borcherds products and the injectivity of the Kudla-Millson theta lift

We prove a converse theorem for the multiplicative Borcherds lift for lattices of square-free level whose associated discriminant group is anisotropic. This can be seen as generalization of Bruinier's results in \cite{Br2}, which provides a converse theorem for lattices of prime level. The surjectivity of the Borcherds lift in our case follows from the injectivity of the Kudla-Millson theta lift. We generalize the corresponding results in \cite{BF1} to the aforementioned lattices and thereby in particular to lattices which are not unimodular and not of type $(p,2)$. Along the way, we compute the contribution of both, the non-Archimedean and Archimedean places of the $L^2$-norm of the Kudla-Millson theta lift. As an application we refine a theorem of Scheithauer on the non-existence of reflective automorphic products.

math.NT↗
arXiv2023

Learning to Do or Learning While Doing: Reinforcement Learning and Bayesian Optimisation for Online Continuous Tuning

Online tuning of real-world plants is a complex optimisation problem that continues to require manual intervention by experienced human operators. Autonomous tuning is a rapidly expanding field of research, where learning-based methods, such as Reinforcement Learning-trained Optimisation (RLO) and Bayesian optimisation (BO), hold great promise for achieving outstanding plant performance and reducing tuning times. Which algorithm to choose in different scenarios, however, remains an open question. Here we present a comparative study using a routine task in a real particle accelerator as an example, showing that RLO generally outperforms BO, but is not always the best choice. Based on the study's results, we provide a clear set of criteria to guide the choice of algorithm for a given tuning task. These can ease the adoption of learning-based autonomous tuning solutions to the operation of complex real-world plants, ultimately improving the availability and pushing the limits of operability of these facilities, thereby enabling scientific and engineering advancements.

cs.LG↗
arXiv2021

On analytic properties of the standard zeta function attached to a vector valued modular form

We proof a Garrett-Böcherer decomposition of a vector valued Siegel Eisenstein series $E_{l,0}^2$ of genus 2 transforming with the Weil representation of $\text{Sp}_2(\mathbb{Z})$ on the group ring $\mathbb{C}[(L'/L)^2]$. We show that the standard zeta function associated to a vector valued common eigenform $f$ for the Weil representation can be meromorphically continued to the whole $s$-plane and that it satisfies a functional equation. The proof is based on an integral representation of this zeta function in terms of $f$ and $E_{l,0}^2$.

math.NT↗
arXiv2021

On vector valued automorphic forms for the Weil representation

We develop a theory of vector valued automorphic forms associated to the Weil representation $\omega_f$ and corresponding to vector valued modular forms transforming with the ``finite'' Weil representation $\rho_L$. For each prime $p$ we determine the structure of a vector valued spherical Hecke algebra depending on $\omega_f$, which acts on the space of automorphic forms.

math.NT↗
arXiv2020

Smart Resource Management for Data Streaming using an Online Bin-packing Strategy

Data stream processing frameworks provide reliable and efficient mechanisms for executing complex workflows over large datasets. A common challenge for the majority of currently available streaming frameworks is efficient utilization of resources. Most frameworks use static or semi-static settings for resource utilization that work well for established use cases but lead to marginal improvements for unseen scenarios. Another pressing issue is the efficient processing of large individual objects such as images and matrices typical for scientific datasets. HarmonicIO has proven to be a good solution for streams of relatively large individual objects, as demonstrated in a benchmark comparison with the Spark and Kafka streaming frameworks. We here present an extension of the HarmonicIO framework based on the online bin-packing algorithm, to allow for efficient utilization of resources. Based on a real world use case from large-scale microscopy pipelines, we compare results of the new system to Spark's auto-scaling mechanism.

cs.DC↗
arXiv2019

An Adaptive Sample Size Trust-Region Method for Finite-Sum Minimization

We propose a trust-region method for finite-sum minimization with an adaptive sample size adjustment technique, which is practical in the sense that it leads to a globally convergent method that shows strong performance empirically without the need for experimentation by the user. During the optimization process, the size of the samples is adaptively increased (or decreased) depending on the progress made on the objective function. We prove that after a finite number iterations the sample includes all points from the data set and the method becomes a full-batch trust-region method. Numerical experiments on convex and nonconvex problems support our claim that our algorithm has significant advantages compared to current state-of-the-art methods.

math.OC↗
arXiv2016

On Globally Diffeomorphic Polynomial Maps via Newton Polytopes and Circuit Numbers

In this article we analyze the global diffeomorphism property of polynomial maps $F:\mathbb{R}^n\rightarrow\mathbb{R}^n$ by studying the properties of the Newton polytopes at infinity corresponding to the sum of squares polynomials $\|F\|_2^2$. This allows us to identify a class of polynomial maps $F$ for which their global diffeomorphism property on $\mathbb{R}^n$ is equivalent to their Jacobian determinant $\text{det }JF$ vanishing nowhere on $\mathbb{R}^n$. In other words, we identify a class of polynomial maps for which the Real Jacobian Conjecture, which was proven to be false in general, still holds.

math.AG↗