Searcharxiv⌕ Search

arXiv subjects

Stefan Schiffer

Publications and source records attributed to Stefan Schiffer.

17 recordsLinked to original sources

Higher integrability of solutions to elliptic equations under additional sign constraints

Solutions to elliptic equations often exhibit higher regularity properties such as \emph{higher integrability}. That is, for instance, a solution $u$ to a system that a priori only satisfies $ u \in W^{1,r}$ is more regular and even in the Sobolev space $W^{1,s}$ for some $s>r$. Under additional constraints of the sign of specific terms such as $(\partial_i u)$ this improvement of regularity can be sharpened further. In this work, we consider two examples of such higher integrability results: First, we show a version of Müller's result on the higher integrability of the determinant for maps $u \in W^{1,n} $ such that $\mathrm{det}(\nabla u) \geq 0$ (or $ \mathrm{det}_-(\nabla u) \in L \log L$). Second, we consider (very weak) solutions to the $p$-Laplace equation that satisfy sign constraints for their partial derivatives, i.e. that $(\partial_i u)_- $ is of higher integrability than $(\partial_i u)_+$. To prove our results, we use the method of Lipschitz truncation; for the second example we further develop a variation of this technique, the \emph{asymmetric} Lipschitz truncation.

math.AP↗

Look: AI at Work! -- Analysing Key Aspects of AI-support at the Work Place

In this paper we present an analysis of technological and psychological factors of applying artificial intelligence (AI) at the work place. We do so for a number of twelve application cases in the context of a project where AI is integrated at work places and in work systems of the future. From a technological point of view we mainly look at the areas of AI that the applications are concerned with. This allows to formulate recommendations in terms of what to look at in developing an AI application and what to pay attention to with regards to building AI literacy with different stakeholders using the system. This includes the importance of high-quality data for training learning-based systems as well as the integration of human expertise, especially with knowledge-based systems. In terms of the psychological factors we derive research questions to investigate in the development of AI supported work systems and to consider in future work, mainly concerned with topics such as acceptance, openness, and trust in an AI system.

cs.HC↗

Insights from Interviews with Teachers and Students on the Use of a Social Robot in Computer Science Class in Sixth Grade

In this paper we report on first insights from interviews with teachers and students on using social robots in computer science class in sixth grade. Our focus is on learning about requirements and potential applications. We are particularly interested in getting both perspectives, the teachers' and the learners' view on how robots could be used and what features they should or should not have. Results show that teachers as well as students are very open to robots in the classroom. However, requirements are partially quite heterogeneous among the groups. This leads to complex design challenges which we discuss at the end of this paper.

cs.RO↗

A variational view on constitutive laws in parabolic problems

We consider a variational approach to solve parabolic problems by minimising a functional over time and space. To achieve existence results we investigate the notion of $\mathscr{A}$-quasiconvexity for non-homogeneous operators in anisotropic spaces. The abstract theory is then applied to formulate a variational solution concept for the non-Newtonian Navier--Stokes equations.

math.AP↗

A variational approach to the Navier-Stokes equations with shear-dependent viscosity

We present a variational approach for the construction of Leray-Hopf solutions to the non-Newtonian Navier-Stokes system. Inspired by the work [42] on the corresponding Newtonian problem, we minimise certain stabilised Weighted Inertia-Dissipation-Energy (WIDE) functionals and pass to the limit of a vanishing parameter in order to recover a Leray-Hopf solution of the non-Newtonian Navier-Stokes equations. The investigation of the non-Newtonian Navier-Stokes system via this variational approach is particularly well suited to gain insights into weak, respectively strong convergence properties of approximating sequences for different flow-behaviour exponents. With this analysis we extend the results of [4] to power-law exponents $\tfrac{2d}{d+2} < p < \tfrac{3d+2}{d+2}$, where weak solutions do not satisfy the energy equality and the involved convergence is genuinely weak. Key of the argument is to pass to the limit in the nonlinear viscosity term in the time-dependent setting. For this we provide an elliptic-parabolic solenoidal Lipschitz truncation that might be of independent interest.

math.AP↗

On incompressible flows in discrete networks and Shnirelman's inequality

Let $f$ and $g$ be two volume-preserving diffeomorphisms on the cube $Q=[0,1]^ν$, $ν\geq 3$. We show that there is a divergence-free vector field $v \in L^1((0,1);L^p(Q))$ such that $v$ connects $f$ and $g$ through the corresponding flow and $\Vert v \Vert_{L^1_t L^p_x} \leq C_{p,ν} \Vert f- g \Vert_{L^p_x}$. In particular we show Shnirelman's inequality, cf. [Shnirelman, Generalized fluid flows, their approximation and applications (1994)], for the optimal Hölder exponent $α=1$, thus proving that the metric on the group of volume-preserving diffeomorphisms of $Q$ is equivalent to the $L^2$-distance. To achieve this, we discretise our problem, use some results on flows in discrete networks and then construct a flow in non-discrete space-time out of the discrete solution.

math.AP↗

$\mathscr{A}$-free truncation and higher integrability of minimisers

We show higher integrability of minimisers of functionals \[ I(u) = \int_Ω f(x,u(x)) ~\mathrm{d}x \] subject to a differential constraint $\mathscr{A} u=0$ under natural $p$-growth and $p$-coercivity conditions for $f$ and regularity assumptions on $Ω$. For the differential operator $\mathscr{A}$ we asssume a rather abstract truncation property that, for instance, holds for operators $\mathscr{A}=\mathrm{curl}$ and $\mathscr{A}=\mathrm{div}$. The proofs are based on the comparison of the minimiser to the truncated version of the minimiser.

math.AP↗

Extensions of divergence-free fields in $\mathrm{L}^{1}$-based function spaces

We establish the first extension results for divergence-free (or solenoidal) elements of $\mathrm{L}^{1}$-based function spaces. Here, the key point is to preserve the solenoidality constraint while simultaneously keeping the underlying $\mathrm{L}^{1}$-boundedness. While previous results as in Kato et al. [Extension and representation of divergence-free vector fields on bounded domains, Math. Res. Lett., 2000] for $\mathrm{L}^{p}$-based function spaces, $1<p<\infty$, rely on PDE approaches, basic principles from harmonic analysis rule out such strategies in the $\mathrm{L}^{1}$-context. By means of a novel method adapted to the divergence-free constraint via differential forms, we establish the existence of such extension operators in the $\mathrm{L}^{1}$-based situation. This applies both to the case of convex domains, where a global extensions can be achieved, as well as to the Lipschitz case, where a local extension can be achieved. Being applicable to $1<p<\infty$ too, our method provides a unifying approach to the cases $p\in\{1,\infty\}$ and $1<p<\infty$. Specifically, covering the exponents $p\in\{1,\infty\}$, this answers a borderline case left open by Kato et al. [Extension and representation of divergence-free vector fields on bounded domains, Math. Res. Lett., 2000] in the affirmative. By use of explicit examples, the assumptions on the underlying domains are shown to be almost optimal.

math.AP↗

BUSSARD -- Better Understanding Social Situations for Autonomous Robot Decision-Making

We report on our effort to create a corpus dataset of different social context situations in an office setting for further disciplinary and interdisciplinary research in computer vision, psychology, and human-robot-interaction. For social robots to be able to behave appropriately, they need to be aware of the social context they act in. Consider, for example, a robot with the task to deliver a personal message to a person. If the person is arguing with an office mate at the time of message delivery, it might be more appropriate to delay playing the message as to respect the recipient's privacy and not to interfere with the current situation. This can only be done if the situation is classified correctly and in a second step if an appropriate behavior is chosen that fits the social situation. Our work aims to enable robots accomplishing the task of classifying social situations by creating a dataset composed of semantically annotated video scenes of office situations from television soap operas. The dataset can then serve as a basis for conducting research in both computer vision and human-robot interaction.

cs.RO↗

Potential Ways to Detect Unfairness in HRI and to Re-establish Positive Group Dynamics

This paper focuses on the identification of different algorithm-based biases in robotic behaviour and their consequences in human-robot mixed groups. We propose to develop computational models to detect episodes of microaggression, discrimination, and social exclusion informed by a) observing human coping behaviours that are used to regain social inclusion and b) using system inherent information that reveal unequal treatment of human interactants. Based on this information we can start to develop regulatory mechanisms to promote fairness and social inclusion in HRI.

cs.RO↗

On the complex constant rank condition and inequalities for differential operators

In this note, we study the complex constant rank condition for differential operators and its implications for coercive differential inequalities. These are inequalities of the form \[ \Vert A u \Vert_{L^p} \leq \Vert \mathscr{A} u \Vert_{L^q}, \] for exponents $1\leq p,q <\infty$ and homogeneous constant-coefficient differential operators $A$ and $\mathscr{A}$. The functions $u \colon Ω\to \mathbb{R}^d$ are defined on open and bounded sets $Ω\subset \mathbb{R}^N$ satisfying certain regularity assumptions. Depending on the order of $A$ and $\mathscr{A}$, such an inequality might be viewed as a generalisation of either Korn's or Sobolev's inequality, respectively. In both cases, as we are on bounded domains, we assume that the Fourier symbol of $\mathscr{A}$ satisfies an algebraic condition, the complex constant rank property.

math.AP↗

An alternative approach to solenoidal Lipschitz truncation

In this work, a new approach to obtain a solenoidal Lipschitz truncation is presented. More precisely, the goal of the truncation is to modify a function $u \in W^{1,p}(\mathbb{R}^3,\mathbb{R}^3)$ that satisfies the additional constraint $\mathrm{div}~ u=0$, such that its modification $\tilde{u}$ is in $W^{1,\infty}(\mathbb{R}^3,\mathbb{R}^3)$ and still is divergence-free. We give an alternative approach to Lipschitz truncation compared to previous works by Breit, Diening & Fuchs (2012) and Breit, Diening & Schwarzacher (2013). The ansatz pursued here allows a rather strict bound on the $W^{1,p}$ distance of $u$ and $\tilde{u}$.

math.AP↗

A data-driven approach to viscous fluid mechanics -- the stationary case

We introduce a data-driven approach to the modelling and analysis of viscous fluid mechanics. Instead of including constitutive laws for the fluid's viscosity in the mathematical model, we suggest to directly use experimental data. Only a set of differential constraints, derived from first principles, and boundary conditions are kept of the classical PDE model and are combined with a data set. The mathematical framework builds on the recently introduced data-driven approach to solid-mechanics [KO16,CMO18]. We construct optimal data-driven solutions that are material model free in the sense that no assumptions on the rheological behaviour of the fluid are made or extrapolated from the data. The differential constraints of fluid mechanics are recast in the language of constant rank differential operators. Adapting abstract results on lower-semicontinuity and $\mathscr{A}$-quasiconvexity, we show a $Γ$-convergence result for the functionals arising in the data-driven fluid mechanical problem. The theory is extended to compact nonlinear perturbations, whence our results apply to both inertialess fluids and flows with finite Reynolds number. Data-driven solutions provide a new relaxed solution concept. We prove that the constructed data-driven solutions are consistent with solutions to the classical PDEs of fluid mechanics if the data sets have the form of a monotone constitutive relation.

math.AP↗

Natural annihilators and operators of constant rank over $\mathbb{C}$

Even if the Fourier symbols of two constant rank differential operators have the same nullspace for each non-trivial phase space variable, the nullspaces of those differential operators might differ by an infinite dimensional space. Under the natural condition of constant rank over $\mathbb{C}$, we establish that the equality of nullspaces on the Fourier symbol level already implies the equality of the nullspaces of the differential operators in $\mathscr{D}'$ modulo polynomials of a fixed degree. In particular, this condition allows to speak of natural annihilators within the framework of complexes of differential operators. As an application, we establish a Poincaré-type lemma for differential operators of constant complex rank in two dimensions.

math.AP↗

A sufficient and necessary condition for $\mathcal{A}$-quasiaffinity

We consider a homogeneous, constant rank differential operator $\mathcal{A}$ and prove a characterisation theorem for $\mathcal{A}$-quasiaffine functions in the spirit of Ball, Currie and Olver (1981); i.e. functions such that \[ f(v) = \int_{T_N} f(v + ψ(y))~\mathrm{d}y \] for all $v$ and all $\mathcal{A}$-free test functions $ψ$ with zero mean. This result is used to get a sufficient, but not necessary condition for the differential operator $\mathcal{A}$, such that linearity along the characteristic cone of $\mathcal{A}$ implies $\mathcal{A}$-quasiaffinity. We show that this implication is true if $\mathcal{A}$ admits a first order potential.

math.AP↗

On symmetric div-quasiconvex hulls and divsym-free $\mathrm{L}^\infty$-truncations

We establish that for any non-empty, compact set $K\subset\mathbb{R}_{\mathrm{sym}}^{3\times 3}$ the $1$- and $\infty$-symmetric div-quasiconvex hulls $K^{(1)}$ and $K^{(\infty)}$ coincide. This settles a conjecture in a recent work of Conti, Müller and Ortiz (Symmetric Div-Quasiconvexity and the Relaxation of Static Problems. Arch. Ration. Mech. Anal. 235(2):841-880) in the affirmative. As a key novelty, we construct an $\mathrm{L}^{\infty}$-truncation that preserves both symmetry and solenoidality of matrix-valued maps in $\mathrm{L}^{1}$.

math.AP↗

$L^{\infty}$-truncation of closed differential forms

In this paper, we prove that for each closed differential form $u \in L^1(\mathbb{R}^N,(\mathbb{R}^N)^{\ast} \wedge ... \wedge (\mathbb{R}^N)^{\ast})$, which is almost in $L^{\infty}$ in the sense that \[ \int_{\{y \in \mathbb{R}^N \colon \vert u(y) \vert \geq L \}} \vert u(y) \vert dy < \varepsilon \] for some $L>0$ and a small $\varepsilon >0$, we may find a closed differential form $v$, such that $\Vert u - v \Vert_{L^1}$ is again small, and $v$ is, in addition, in $L^{\infty}$ with a bound on its $L^{\infty}$ norm depending only on $N$ and $L$. In particular, the set $\{ v \neq u\}$ has measure at most $C \varepsilon$. We then look at applications of this theorem. We are able to prove that the $\mathcal{A}$-$p$-quasiconvex hull of a set does not depend on $p$. Furthermore, we can prove a classification theorem for $\mathcal{A}$-$\infty$-Young measures.

math.AP↗