-
Forbidden subgraphs of conjugacy class graphs of groups
Authors:
Papi Ray,
Sonakshee Arora,
Pallabi Manna
Abstract:
Let G be a finite group. The nilpotent/commuting/solvable conjugacy class graph is a simple graph with non-central conjugacy classes of G as its vertex set and two vertices are adjacent if and only if a member of one conjugacy class with a member of another conjugacy class generates a nilpotent/abelian/solvable (sub)group. In this paper have discussed about the forbidden subgraphs of nilpotent, co…
▽ More
Let G be a finite group. The nilpotent/commuting/solvable conjugacy class graph is a simple graph with non-central conjugacy classes of G as its vertex set and two vertices are adjacent if and only if a member of one conjugacy class with a member of another conjugacy class generates a nilpotent/abelian/solvable (sub)group. In this paper have discussed about the forbidden subgraphs of nilpotent, commuting and solvable conjugacy class graphs of groups.
△ Less
Submitted 3 June, 2024;
originally announced June 2024.
-
Eventually positive semigroups: spectral and asymptotic analysis
Authors:
Sahiba Arora
Abstract:
The spectral theory of semigroup generators is a crucial tool for analysing the asymptotic properties of operator semigroups. Typically, Tauberian theorems, such as the ABLV theorem, demand extensive information about the spectrum to derive convergence results. However, the scenario is significantly simplified for positive semigroups on Banach lattices. This observation extends to the broader clas…
▽ More
The spectral theory of semigroup generators is a crucial tool for analysing the asymptotic properties of operator semigroups. Typically, Tauberian theorems, such as the ABLV theorem, demand extensive information about the spectrum to derive convergence results. However, the scenario is significantly simplified for positive semigroups on Banach lattices. This observation extends to the broader class of eventually positive semigroups -- a phenomenon observed in various concrete differential equations. In this paper, we investigate the spectral and asymptotic properties of eventually positive semigroups, focusing particularly on the persistently irreducible case. Our findings expand upon the existing theory of eventual positivity, offering new insights into the cyclicity of the peripheral spectrum and asymptotic trends. Notably, several arguments for positive operators and semigroups do not apply in our context, necessitating the use of ultrapower arguments to circumvent these challenges.
△ Less
Submitted 25 May, 2024;
originally announced May 2024.
-
The lattice structure of negative Sobolev and extrapolation spaces
Authors:
Sahiba Arora,
Jochen Glück,
Felix L. Schwenninger
Abstract:
It is well-known that the Sobolev spaces $W^{k,p}(\mathbb R^d)$ are vector lattices with respect to the pointwise almost everywhere order if $k \in \{0,1\}$, but not if $k \ge 2$. In this note, we consider negative $k$ and show that the span of the positive cone in $W^{k,p}(\mathbb R^d)$ is a vector lattice in this case.
We also prove a related abstract result: if $(T(t))_{t \in [0,\infty)}$ is…
▽ More
It is well-known that the Sobolev spaces $W^{k,p}(\mathbb R^d)$ are vector lattices with respect to the pointwise almost everywhere order if $k \in \{0,1\}$, but not if $k \ge 2$. In this note, we consider negative $k$ and show that the span of the positive cone in $W^{k,p}(\mathbb R^d)$ is a vector lattice in this case.
We also prove a related abstract result: if $(T(t))_{t \in [0,\infty)}$ is a positive $C_0$-semigroup on a Banach lattice $X$ with order continuous norm, then the span of the cone $X_{-1,+}$ in the extrapolation space $X_{-1}$ is a vector lattice. This complements results obtained by Bátkai, Jacob, Wintermayr, and Voigt in the context of perturbation theory and provides additional context for the theory of infinite-dimensional positive systems.
△ Less
Submitted 2 April, 2024;
originally announced April 2024.
-
Limit-case admissibility for positive infinite-dimensional systems
Authors:
Sahiba Arora,
Jochen Glück,
Lassi Paunonen,
Felix L. Schwenninger
Abstract:
In the context of positive infinite-dimensional linear systems, we systematically study $L^p$-admissible control and observation operators with respect to the limit-cases $p=\infty$ and $p=1$, respectively. This requires an in-depth understanding of the order structure on the extrapolation space $X_{-1}$, which we provide. These properties of $X_{-1}$ also enable us to discuss when zero-class admi…
▽ More
In the context of positive infinite-dimensional linear systems, we systematically study $L^p$-admissible control and observation operators with respect to the limit-cases $p=\infty$ and $p=1$, respectively. This requires an in-depth understanding of the order structure on the extrapolation space $X_{-1}$, which we provide. These properties of $X_{-1}$ also enable us to discuss when zero-class admissibility is automatic. While those limit-cases are the weakest form of admissibility on the $L^p$-scale, it is remarkable that they sometimes follow from order theoretic and geometric assumptions. Our assumptions on the geometries of the involved spaces are minimal.
△ Less
Submitted 4 May, 2024; v1 submitted 1 April, 2024;
originally announced April 2024.
-
Domination of semigroups on standard forms of von Neumann algebras
Authors:
Sahiba Arora,
Ralph Chill,
Sachi Srivastava
Abstract:
Consider $(T_t)_{t\ge 0}$ and $(S_t)_{t\ge 0}$ as real $C_0$-semigroups generated by closed and symmetric sesquilinear forms on a standard form of a von Neumann algebra. We provide a characterisation for the domination of the semigroup $(T_t)_{t\ge 0}$ by $(S_t)_{t\ge 0}$, which means that $-S_t v\le T_t u\le S_t v$ holds for all $t\ge 0$ and all real $u$ and $v$ that satisfy $-v\le u\le v$. This…
▽ More
Consider $(T_t)_{t\ge 0}$ and $(S_t)_{t\ge 0}$ as real $C_0$-semigroups generated by closed and symmetric sesquilinear forms on a standard form of a von Neumann algebra. We provide a characterisation for the domination of the semigroup $(T_t)_{t\ge 0}$ by $(S_t)_{t\ge 0}$, which means that $-S_t v\le T_t u\le S_t v$ holds for all $t\ge 0$ and all real $u$ and $v$ that satisfy $-v\le u\le v$. This characterisation extends the Ouhabaz characterisation for semigroup domination to the non-commutative $L^2$ spaces. Additionally, we present a simpler characterisation when both semigroups are positive as well as consider the setting in which $(T_t)_{t\ge 0}$ need not be real.
△ Less
Submitted 14 November, 2023; v1 submitted 5 September, 2023;
originally announced September 2023.
-
Irreducibility of eventually positive semigroups
Authors:
Sahiba Arora,
Jochen Glück
Abstract:
Positive $C_0$-semigroups that occur in concrete applications are, more often than not, irreducible. Therefore a deep and extensive theory of irreducibility has been developed that includes characterizations, perturbation analysis, and spectral results. Many arguments from this theory, however, break down if the semigroup is only eventually positive - a property that has recently been shown to occ…
▽ More
Positive $C_0$-semigroups that occur in concrete applications are, more often than not, irreducible. Therefore a deep and extensive theory of irreducibility has been developed that includes characterizations, perturbation analysis, and spectral results. Many arguments from this theory, however, break down if the semigroup is only eventually positive - a property that has recently been shown to occur in numerous concrete evolution equations.
In this article, we develop new tools that also work for the eventually positive case. The lack of positivity for small times makes it necessary to consider ideals that might only be invariant for large times. In contrast to their classical counterparts - the invariant ideals - such eventually invariant ideals require more involved methods from the theory of operator ranges.
Using those methods we are able to characterize (an appropriate adaptation of) irreducibility by means of linear functionals, derive a perturbation result, prove a number of spectral theorems, and analyze the interaction of irreducibility with analyticity, all in the eventually positive case. By a number of examples, we illustrate what kind of behaviour can and cannot be expected in this setting.
△ Less
Submitted 10 June, 2024; v1 submitted 10 July, 2023;
originally announced July 2023.
-
Implicit Bias of Gradient Descent on Reparametrized Models: On Equivalence to Mirror Descent
Authors:
Zhiyuan Li,
Tianhao Wang,
JasonD. Lee,
Sanjeev Arora
Abstract:
As part of the effort to understand implicit bias of gradient descent in overparametrized models, several results have shown how the training trajectory on the overparametrized model can be understood as mirror descent on a different objective. The main result here is a characterization of this phenomenon under a notion termed commuting parametrization, which encompasses all the previous results i…
▽ More
As part of the effort to understand implicit bias of gradient descent in overparametrized models, several results have shown how the training trajectory on the overparametrized model can be understood as mirror descent on a different objective. The main result here is a characterization of this phenomenon under a notion termed commuting parametrization, which encompasses all the previous results in this setting. It is shown that gradient flow with any commuting parametrization is equivalent to continuous mirror descent with a related Legendre function. Conversely, continuous mirror descent with any Legendre function can be viewed as gradient flow with a related commuting parametrization. The latter result relies upon Nash's embedding theorem.
△ Less
Submitted 8 July, 2022;
originally announced July 2022.
-
Topological groups with a compact open subgroup, Relative hyperbolicity and Coherence
Authors:
Shivam Arora,
Eduardo Martínez-Pedroza
Abstract:
The main objects of study in this article are pairs $(G, \mathcal{H})$ where $G$ is a topological group with a compact open subgroup, and $\mathcal{H}$ is a finite collection of open subgroups. We develop geometric techniques to study the notions of $G$ being compactly generated and compactly presented relative to $\mathcal H$. This includes topological characterizations in terms of discrete actio…
▽ More
The main objects of study in this article are pairs $(G, \mathcal{H})$ where $G$ is a topological group with a compact open subgroup, and $\mathcal{H}$ is a finite collection of open subgroups. We develop geometric techniques to study the notions of $G$ being compactly generated and compactly presented relative to $\mathcal H$. This includes topological characterizations in terms of discrete actions of $G$ on complexes, quasi-isometry invariance of certain graphs associated to the pairs $(G,\mathcal H)$ when $G$ is compactly generated relative to $\mathcal H$, and extensions of known results for the discrete case. For example, generalizing results of Osin for discrete groups, we show that in the case that $G$ is compactly presented relative to $\mathcal H$:
$\bullet$ if $G$ is compactly generated, then each subgroup $H\in \mathcal H$ is compactly generated;
$\bullet$ if each subgroup $H\in \mathcal H$ is compactly presented, then $G$ is compactly presented.
The article also introduces an approach to relative hyperbolicity for pairs $(G, \mathcal H)$ based on Bowditch's work using discrete actions on hyperbolic fine graphs. For example, we prove that if $G$ is hyperbolic relative to $\mathcal H$ then $G$ is compactly presented relative to $\mathcal H$. As applications of the results of the article we prove combination results for coherent topological groups with a compact open subgroup, and extend McCammond-Wise perimeter method to this general framework.
△ Less
Submitted 9 May, 2023; v1 submitted 14 June, 2022;
originally announced June 2022.
-
Understanding Gradient Descent on Edge of Stability in Deep Learning
Authors:
Sanjeev Arora,
Zhiyuan Li,
Abhishek Panigrahi
Abstract:
Deep learning experiments by Cohen et al. [2021] using deterministic Gradient Descent (GD) revealed an Edge of Stability (EoS) phase when learning rate (LR) and sharpness (i.e., the largest eigenvalue of Hessian) no longer behave as in traditional optimization. Sharpness stabilizes around $2/$LR and loss goes up and down across iterations, yet still with an overall downward trend. The current pape…
▽ More
Deep learning experiments by Cohen et al. [2021] using deterministic Gradient Descent (GD) revealed an Edge of Stability (EoS) phase when learning rate (LR) and sharpness (i.e., the largest eigenvalue of Hessian) no longer behave as in traditional optimization. Sharpness stabilizes around $2/$LR and loss goes up and down across iterations, yet still with an overall downward trend. The current paper mathematically analyzes a new mechanism of implicit regularization in the EoS phase, whereby GD updates due to non-smooth loss landscape turn out to evolve along some deterministic flow on the manifold of minimum loss. This is in contrast to many previous results about implicit bias either relying on infinitesimal updates or noise in gradient. Formally, for any smooth function $L$ with certain regularity condition, this effect is demonstrated for (1) Normalized GD, i.e., GD with a varying LR $η_t =\fracη{\| \nabla L(x(t)) \|}$ and loss $L$; (2) GD with constant LR and loss $\sqrt{L- \min_x L(x)}$. Both provably enter the Edge of Stability, with the associated flow on the manifold minimizing $λ_{1}(\nabla^2 L)$. The above theoretical results have been corroborated by an experimental study.
△ Less
Submitted 28 October, 2022; v1 submitted 19 May, 2022;
originally announced May 2022.
-
Criteria for eventual domination of operator semigroups and resolvents
Authors:
Sahiba Arora,
Jochen Glück
Abstract:
We consider two $C_0$-semigroups $(e^{tA})_{t \ge 0}$ and $(e^{tB})_{t \ge 0}$ on function spaces (or, more generally, on Banach lattices) and analyse eventual domination between them in the sense that $|e^{tA}f| \le e^{tB}|f|$ for all sufficiently large times $t$. We characterise this behaviour and prove a number of theoretical results which complement earlier results given by Mugnolo and the sec…
▽ More
We consider two $C_0$-semigroups $(e^{tA})_{t \ge 0}$ and $(e^{tB})_{t \ge 0}$ on function spaces (or, more generally, on Banach lattices) and analyse eventual domination between them in the sense that $|e^{tA}f| \le e^{tB}|f|$ for all sufficiently large times $t$. We characterise this behaviour and prove a number of theoretical results which complement earlier results given by Mugnolo and the second author in the special case where both semigroups are positive for large times.
Moreover, we study the analogous question of whether the resolvent of $B$ eventually dominates the resolvent of $A$ close to the spectral bound of $B$. This is closely related to the so-called maximum and anti-maximum principles. In order to demonstrate how our results can be used, we include several applications to concrete differential operators.
At the end of the paper, we demonstrate that eventual positivity of the resolvent of a semigroup generator is closely related to eventual positivity of the Cesàro means of the associated semigroup.
△ Less
Submitted 17 December, 2022; v1 submitted 31 March, 2022;
originally announced April 2022.
-
A characterization of the individual maximum and anti-maximum principle
Authors:
Sahiba Arora,
Jochen Glück
Abstract:
Abstract approaches to maximum and anti-maximum principles for differential operators typically rely on the condition that all vectors in the domain of the operator are dominated by the leading eigenfunction of the operator. We study the necessity of this condition. In particular, we show that under a number of natural assumptions, so-called individual versions of both the maximum and the anti-max…
▽ More
Abstract approaches to maximum and anti-maximum principles for differential operators typically rely on the condition that all vectors in the domain of the operator are dominated by the leading eigenfunction of the operator. We study the necessity of this condition. In particular, we show that under a number of natural assumptions, so-called individual versions of both the maximum and the anti-maximum principle simultaneously hold if and only if the aforementioned domination condition is satisfied.
Consequently, we are able to show that a variety of concrete differential operators do not satisfy an anti-maximum principle.
△ Less
Submitted 28 August, 2023; v1 submitted 10 March, 2022;
originally announced March 2022.
-
Adaptive Gradient Methods with Local Guarantees
Authors:
Zhou Lu,
Wenhan Xia,
Sanjeev Arora,
Elad Hazan
Abstract:
Adaptive gradient methods are the method of choice for optimization in machine learning and used to train the largest deep models. In this paper we study the problem of learning a local preconditioner, that can change as the data is changing along the optimization trajectory. We propose an adaptive gradient method that has provable adaptive regret guarantees vs. the best local preconditioner. To d…
▽ More
Adaptive gradient methods are the method of choice for optimization in machine learning and used to train the largest deep models. In this paper we study the problem of learning a local preconditioner, that can change as the data is changing along the optimization trajectory. We propose an adaptive gradient method that has provable adaptive regret guarantees vs. the best local preconditioner. To derive this guarantee, we prove a new adaptive regret bound in online learning that improves upon previous adaptive online learning methods. We demonstrate the robustness of our method in automatically choosing the optimal learning rate schedule for popular benchmarking tasks in vision and language domains. Without the need to manually tune a learning rate schedule, our method can, in a single run, achieve comparable and stable task accuracy as a fine-tuned optimizer.
△ Less
Submitted 25 January, 2023; v1 submitted 2 March, 2022;
originally announced March 2022.
-
Total Controllability of nonlocal semilinear functional evolution equations with non-instantaneous impulses
Authors:
J. Kumar,
S. Singh,
S. Arora,
J. Dabas
Abstract:
In this article, we are discussing a more vital concept of controllability, termed total controllability. We have considered a nonlocal semilinear functional evolution equations with non-instantaneous impulses and finite delay in Hilbert spaces. A set of sufficient conditions of total controllability is obtained for the evolution system under consideration, by imposing the theory of C_0-semigroup…
▽ More
In this article, we are discussing a more vital concept of controllability, termed total controllability. We have considered a nonlocal semilinear functional evolution equations with non-instantaneous impulses and finite delay in Hilbert spaces. A set of sufficient conditions of total controllability is obtained for the evolution system under consideration, by imposing the theory of C_0-semigroup and Banach fixed point theorem. We also established the total controllability results for a functional integro-differential equation. Finally, an example is given to demonstrate the feasibility of derived abstract results.
△ Less
Submitted 3 January, 2022;
originally announced January 2022.
-
Approximate controllability of non-autonomous second order impulsive functional evolution equations in Banach spaces
Authors:
Sumit Arora,
Soniya Singh,
Manil T. Mohan,
Jaydev Dabas
Abstract:
This article investigates the approximate controllability of second order non-autonomous functional evolution equations involving non-instantaneous impulses and nonlocal conditions. First, we discuss the approximate controllability of second order linear system in detail, which lacks in the existing literature. Then, we derive sufficient conditions for approximate controllability of our system in…
▽ More
This article investigates the approximate controllability of second order non-autonomous functional evolution equations involving non-instantaneous impulses and nonlocal conditions. First, we discuss the approximate controllability of second order linear system in detail, which lacks in the existing literature. Then, we derive sufficient conditions for approximate controllability of our system in separable reflexive Banach spaces via linear evolution operator, resolvent operator conditions, and Schauder's fixed point theorem. Moreover, in this paper, we define proper identification of resolvent operator in Banach spaces. Finally, we verify our results to examine the approximate controllability of the non-autonomous wave equation with non-instantaneous impulses and finite delay in the application section.
△ Less
Submitted 31 December, 2021;
originally announced December 2021.
-
Domination of semigroups generated by regular forms
Authors:
Sahiba Arora,
Ralph Chill,
Jean-Daniel Djida
Abstract:
We give a representation for regular forms associated with dominated $C_0$-semigroups which, in turn, characterises domination of $C_0$-semigroups associated with regular forms. In addition, we prove a relationship between the positivity of (dominated) $C_0$-semigroups and the locality of the associated forms.
We give a representation for regular forms associated with dominated $C_0$-semigroups which, in turn, characterises domination of $C_0$-semigroups associated with regular forms. In addition, we prove a relationship between the positivity of (dominated) $C_0$-semigroups and the locality of the associated forms.
△ Less
Submitted 18 April, 2023; v1 submitted 30 November, 2021;
originally announced November 2021.
-
Stability of (eventually) positive semigroups on spaces of continuous functions
Authors:
Sahiba Arora,
Jochen Glück
Abstract:
We present a new and very short proof of the fact that, for positive $C_0$-semigroups on spaces of continuous functions, the spectral and the growth bound coincide. Our argument, inspired by an idea of Vogt, makes the role of the underlying space completely transparent and also works if the space does not contain the constant functions - a situation in which all earlier proofs become technically q…
▽ More
We present a new and very short proof of the fact that, for positive $C_0$-semigroups on spaces of continuous functions, the spectral and the growth bound coincide. Our argument, inspired by an idea of Vogt, makes the role of the underlying space completely transparent and also works if the space does not contain the constant functions - a situation in which all earlier proofs become technically quite involved.
We also show how the argument can be adapted to yield the same result for semigroups that are only eventually positive rather than positive.
△ Less
Submitted 14 July, 2022; v1 submitted 9 October, 2021;
originally announced October 2021.
-
Fixed Point Sets in Diagrammatically Reducible Complexes
Authors:
Shivam Arora,
Eduardo Martínez-Pedroza
Abstract:
Let $H$ be a group acting on a simply-connected diagrammatically reducible combinatorial 2-complex $X$ with fine 1-skeleton. If the fixed point set $X^ H$ is non-empty, then it is contractible. Having fine 1-skeleton is a weaker version of being locally finite.
Let $H$ be a group acting on a simply-connected diagrammatically reducible combinatorial 2-complex $X$ with fine 1-skeleton. If the fixed point set $X^ H$ is non-empty, then it is contractible. Having fine 1-skeleton is a weaker version of being locally finite.
△ Less
Submitted 7 July, 2021; v1 submitted 2 July, 2021;
originally announced July 2021.
-
Approximate controllability of non-instantaneous impulsive fractional evolution equations of order $1<α<2$ with state-dependent delay in Banach spaces
Authors:
S. Arora,
Manil T. Mohan,
J. Dabas
Abstract:
The current article examines the approximate controllability problem for non-instantaneous impulsive fractional evolution equations of order $1<α<2$ with state-dependent delay in separable reflexive Banach spaces. In order to establish sufficient conditions for the approximate controllability of our problem, we first formulate the linear-regulator problem and obtain the optimal control in feedback…
▽ More
The current article examines the approximate controllability problem for non-instantaneous impulsive fractional evolution equations of order $1<α<2$ with state-dependent delay in separable reflexive Banach spaces. In order to establish sufficient conditions for the approximate controllability of our problem, we first formulate the linear-regulator problem and obtain the optimal control in feedback form. By using this optimal control, we deduce the approximate controllability of the linear fractional control system of order $1<α<2$. Further, we derive sufficient conditions for the approximate controllability of the nonlinear problem. Finally, we provide a concrete example to validate the efficiency of the derived results.
△ Less
Submitted 29 June, 2021;
originally announced June 2021.
-
Approximate controllability of fractional order non-instantaneous impulsive functional evolution equations with state-dependent delay in Banach spaces
Authors:
S. Arora,
M. T. Mohan,
J. Dabas
Abstract:
The present paper deals with the control problems governed by fractional non-instantaneous impulsive functional evolution equations with state-dependent delay involving Caputo fractional derivatives in Banach spaces. The main objective of this work is to formulate sufficient conditions for the approximate controllability of the considered system in separable reflexive Banach spaces. We have exploi…
▽ More
The present paper deals with the control problems governed by fractional non-instantaneous impulsive functional evolution equations with state-dependent delay involving Caputo fractional derivatives in Banach spaces. The main objective of this work is to formulate sufficient conditions for the approximate controllability of the considered system in separable reflexive Banach spaces. We have exploited the resolvent operator technique and Schauder's fixed point theorem in the proofs to achieve this goal. The approximate controllability of linear system is discussed in detail, which lacks in the existing literature. We also provide an example to illustrate the efficiency of the developed results. Moreover, we point out some shortcomings of the existing works in the context of characterization of mild solution and phase space, and approximate controllability of fractional order impulsive systems in Banach spaces.
△ Less
Submitted 5 June, 2021;
originally announced June 2021.
-
An operator theoretic approach to uniform (anti-)maximum principles
Authors:
Sahiba Arora,
Jochen Glück
Abstract:
Maximum principles and uniform anti-maximum principles are a ubiquitous topic in PDE theory that is closely tied to the Krein--Rutman theorem and kernel estimates for resolvents.
We take up a classical idea of Takáč - to prove (anti-)maximum principles in an abstract operator theoretic framework - and combine it with recent ideas from the theory of eventually positive operator semigroups. This e…
▽ More
Maximum principles and uniform anti-maximum principles are a ubiquitous topic in PDE theory that is closely tied to the Krein--Rutman theorem and kernel estimates for resolvents.
We take up a classical idea of Takáč - to prove (anti-)maximum principles in an abstract operator theoretic framework - and combine it with recent ideas from the theory of eventually positive operator semigroups. This enables us to derive necessary and sufficient conditions for (anti-)maximum principles in a very general setting. Consequently, we are able to either prove or disprove (anti-)maximum principles for a large variety of concrete differential operators. As a bonus, for several operators that are already known to satisfy or to not satisfy anti-maximum principles, our theory gives a very clear and concise explanation of this behaviour.
△ Less
Submitted 10 December, 2021; v1 submitted 25 April, 2021;
originally announced April 2021.
-
Locally Eventually Positive Operator Semigroups
Authors:
Sahiba Arora
Abstract:
We initiate a theory of locally eventually positive operator semigroups on Banach lattices. Intuitively this means: given a positive initial datum, the solution of the corresponding Cauchy problem becomes (and stays) positive in a part of the domain, after a sufficiently large time. A drawback of the present theory of eventually positive $C_0$-semigroups is that it is applicable only when the lead…
▽ More
We initiate a theory of locally eventually positive operator semigroups on Banach lattices. Intuitively this means: given a positive initial datum, the solution of the corresponding Cauchy problem becomes (and stays) positive in a part of the domain, after a sufficiently large time. A drawback of the present theory of eventually positive $C_0$-semigroups is that it is applicable only when the leading eigenvalue of the semigroup generator has a strongly positive eigenvector. We weaken this requirement and give sufficient criteria for individual and uniform local eventual positivity of the semigroup. This allows us to treat a larger class of examples by giving us more freedom on the domain when dealing with function spaces -- for instance, the square of the Laplace operator with Dirichlet boundary conditions on $L^2$ and the Dirichlet bi-Laplacian on $L^p$-spaces. Besides, we establish various spectral and convergence properties of locally eventually positive semigroups.
△ Less
Submitted 27 July, 2021; v1 submitted 27 January, 2021;
originally announced January 2021.
-
Spectrum and convergence of eventually positive operator semigroups
Authors:
Sahiba Arora,
Jochen Glück
Abstract:
An intriguing feature of positive $C_0$-semigroups on function spaces (or more generally on Banach lattices) is that their long-time behaviour is much easier to describe than it is for general semigroups. In particular, the convergence of semigroup operators (strongly or in the operator norm) as time tends to infinity can be characterized by a set of simple spectral and compactness conditions.
I…
▽ More
An intriguing feature of positive $C_0$-semigroups on function spaces (or more generally on Banach lattices) is that their long-time behaviour is much easier to describe than it is for general semigroups. In particular, the convergence of semigroup operators (strongly or in the operator norm) as time tends to infinity can be characterized by a set of simple spectral and compactness conditions.
In the present paper, we show that similar theorems remain true for the larger class of (uniformly) eventually positive semigroups - which recently arose in the study of various concrete differential equations.
A major step in one of our characterizations is to show a version of the famous Niiro-Sawashima theorem for eventually positive operators. Several proofs for positive operators and semigroups do not work in our setting any longer, necessitating different arguments and giving our approach a distinct flavour.
△ Less
Submitted 5 June, 2021; v1 submitted 9 November, 2020;
originally announced November 2020.
-
Existence and approximate controllability of non-autonomous functional impulsive evolution inclusions in Banach spaces
Authors:
Sumit Arora,
Manil T. Mohan,
Jaydev Dabas
Abstract:
In this paper, we are concerned with the approximate controllability results for a class of impulsive functional differential control systems involving time dependent operators in Banach spaces. First, we show the existence of a mild solution for non-autonomous functional impulsive evolution inclusions in separable reflexive Banach spaces with the help of the evolution family and a generalization…
▽ More
In this paper, we are concerned with the approximate controllability results for a class of impulsive functional differential control systems involving time dependent operators in Banach spaces. First, we show the existence of a mild solution for non-autonomous functional impulsive evolution inclusions in separable reflexive Banach spaces with the help of the evolution family and a generalization of the Leray-Schauder fixed point theorem for multi-valued maps. In order to establish sufficient conditions for the approximate controllability of our problem, we first consider a linear-quadratic regulator problem and obtain the optimal control in the feedback form, which contains the resolvent operator consisting of duality map**. With the help of this optimal control, we prove the approximate controllability of the linear system and hence derive sufficient conditions for the approximate controllability of our problem. Moreover, in this paper, we rectify several shortcomings of the related works available in the literature, namely, proper identification of resolvent operator in Banach spaces, characterization of phase space in the presence of impulsive effects and lack of compactness of the operator $h(\cdot)\mapsto \int_{0}^{\cdot}\mathrm{U}(\cdot,s)h(s)\mathrm{d} s : \mathrm{L}^{1}([0,T];\mathbb{Y}) \rightarrow \mathrm{C}([0,T];\mathbb{Y}),$ where $\mathbb{Y}$ is a Banach space and $\mathrm{U}(\cdot,\cdot)$ is the evolution family, etc. Finally, we provide a concrete example to illustrate the efficiency of our results.
△ Less
Submitted 12 October, 2020;
originally announced October 2020.
-
Impact of packet dropouts on the performance of optimal controllers and observers
Authors:
Amanpreet Singh Arora,
Sanand Dilip
Abstract:
We investigate the impact of packet dropouts due to non-idealities in communication networks on the performance of optimally derived controllers and observers in a minimax sense. These packet dropouts are modeled by discrete constrained switching signals via directed graphs. We consider time optimal control and estimation, minimum energy and fuel optimization and LQR problems for systems subject t…
▽ More
We investigate the impact of packet dropouts due to non-idealities in communication networks on the performance of optimally derived controllers and observers in a minimax sense. These packet dropouts are modeled by discrete constrained switching signals via directed graphs. We consider time optimal control and estimation, minimum energy and fuel optimization and LQR problems for systems subject to packet dropouts which turn out to be combinatorial; and compute algorithmically the worst case performance of these optimization problems. To reduce the computational burden in solving these combinatorial problems, a partial order is imposed on the switching signals. Validation of the proposed methods is done over a set of randomly generated systems. Finally, by associating the worst case performance for each optimal control problem to the underlying network, one can optimize among the set of available communication networks.
△ Less
Submitted 21 June, 2020;
originally announced June 2020.
-
A Sample Complexity Separation between Non-Convex and Convex Meta-Learning
Authors:
Nikunj Saunshi,
Yi Zhang,
Mikhail Khodak,
Sanjeev Arora
Abstract:
One popular trend in meta-learning is to learn from many training tasks a common initialization for a gradient-based method that can be used to solve a new task with few samples. The theory of meta-learning is still in its early stages, with several recent learning-theoretic analyses of methods such as Reptile [Nichol et al., 2018] being for convex models. This work shows that convex-case analysis…
▽ More
One popular trend in meta-learning is to learn from many training tasks a common initialization for a gradient-based method that can be used to solve a new task with few samples. The theory of meta-learning is still in its early stages, with several recent learning-theoretic analyses of methods such as Reptile [Nichol et al., 2018] being for convex models. This work shows that convex-case analysis might be insufficient to understand the success of meta-learning, and that even for non-convex models it is important to look inside the optimization black-box, specifically at properties of the optimization trajectory. We construct a simple meta-learning instance that captures the problem of one-dimensional subspace learning. For the convex formulation of linear regression on this instance, we show that the new task sample complexity of any initialization-based meta-learning algorithm is $Ω(d)$, where $d$ is the input dimension. In contrast, for the non-convex formulation of a two layer linear network on the same instance, we show that both Reptile and multi-task representation learning can have new task sample complexity of $\mathcal{O}(1)$, demonstrating a separation from convex meta-learning. Crucially, analyses of the training dynamics of these methods reveal that they can meta-learn the correct subspace onto which the data should be projected.
△ Less
Submitted 25 February, 2020;
originally announced February 2020.
-
Solution branches of nonlinear eigenvalue problems on restricted domains
Authors:
Shane Arora
Abstract:
We extend bifurcation results of nonlinear eigenvalue problems from real Banach spaces to any neighbourhood of a given point. For points of odd multiplicity on these restricted domains, we establish that the component of solutions through the bifurcation point either is unbounded, admits an accumulation point on the boundary, or contains an even number of odd multiplicity points. In the simple mul…
▽ More
We extend bifurcation results of nonlinear eigenvalue problems from real Banach spaces to any neighbourhood of a given point. For points of odd multiplicity on these restricted domains, we establish that the component of solutions through the bifurcation point either is unbounded, admits an accumulation point on the boundary, or contains an even number of odd multiplicity points. In the simple multiplicity case, we show that branches of solutions in the directions of corresponding eigenvectors satisfy similar conditions on our domains.
△ Less
Submitted 17 February, 2020;
originally announced February 2020.
-
Feedback Linearization for Unknown Systems via Reinforcement Learning
Authors:
Tyler Westenbroek,
David Fridovich-Keil,
Eric Mazumdar,
Shreyas Arora,
Valmik Prabhu,
S. Shankar Sastry,
Claire J. Tomlin
Abstract:
We present a novel approach to control design for nonlinear systems which leverages model-free policy optimization techniques to learn a linearizing controller for a physical plant with unknown dynamics. Feedback linearization is a technique from nonlinear control which renders the input-output dynamics of a nonlinear plant \emph{linear} under application of an appropriate feedback controller. Onc…
▽ More
We present a novel approach to control design for nonlinear systems which leverages model-free policy optimization techniques to learn a linearizing controller for a physical plant with unknown dynamics. Feedback linearization is a technique from nonlinear control which renders the input-output dynamics of a nonlinear plant \emph{linear} under application of an appropriate feedback controller. Once a linearizing controller has been constructed, desired output trajectories for the nonlinear plant can be tracked using a variety of linear control techniques. However, the calculation of a linearizing controller requires a precise dynamics model for the system. As a result, model-based approaches for learning exact linearizing controllers generally require a simple, highly structured model of the system with easily identifiable parameters. In contrast, the model-free approach presented in this paper is able to approximate the linearizing controller for the plant using general function approximation architectures. Specifically, we formulate a continuous-time optimization problem over the parameters of a learned linearizing controller whose optima are the set of parameters which best linearize the plant. We derive conditions under which the learning problem is (strongly) convex and provide guarantees which ensure the true linearizing controller for the plant is recovered. We then discuss how model-free policy optimization algorithms can be used to solve a discrete-time approximation to the problem using data collected from the real-world plant. The utility of the framework is demonstrated in simulation and on a real-world robotic platform.
△ Less
Submitted 21 April, 2020; v1 submitted 29 October, 2019;
originally announced October 2019.
-
An alternative approach to Fréchet derivatives
Authors:
Shane Arora,
Hazel Browne,
Daniel Daners
Abstract:
We discuss an alternative approach to Fréchet derivatives on Banach spaces inspired by a characterisation of derivatives due to Carathéodory. The approach allows us to reduce many questions of differentiability to a question of continuity. We demonstrate how that simplifies the theory of differentiation, including the rules of differentiation and the Schwarz Lemma on the symmetry of second order d…
▽ More
We discuss an alternative approach to Fréchet derivatives on Banach spaces inspired by a characterisation of derivatives due to Carathéodory. The approach allows us to reduce many questions of differentiability to a question of continuity. We demonstrate how that simplifies the theory of differentiation, including the rules of differentiation and the Schwarz Lemma on the symmetry of second order derivatives. We also provide a short proof of the differentiable dependence of fixed points in the Banach fixed point theorem.
△ Less
Submitted 25 October, 2019;
originally announced October 2019.
-
Subgroups, hyperbolicity and cohomological dimension for totally disconnected locally compact groups
Authors:
Shivam Arora,
Ilaria Castellano,
Ged Corob Cook,
Eduardo Martínez-Pedroza
Abstract:
This article is part of the program of studying large-scale geometric properties of totally disconnected locally compact groups, TDLC-groups, by analogy with the theory for discrete groups. We provide a characterization of hyperbolic TDLC-groups, in terms of homological isoperimetric inequalities. This characterization is used to prove the main result of the article: for hyperbolic TDLC-groups wit…
▽ More
This article is part of the program of studying large-scale geometric properties of totally disconnected locally compact groups, TDLC-groups, by analogy with the theory for discrete groups. We provide a characterization of hyperbolic TDLC-groups, in terms of homological isoperimetric inequalities. This characterization is used to prove the main result of the article: for hyperbolic TDLC-groups with rational discrete cohomological dimension $\leq 2$, hyperbolicity is inherited by compactly presented closed subgroups. As a consequence, every compactly presented closed subgroup of the automorphism group $\mathrm{Aut}(X)$ of a negatively curved locally finite $2$-dimensional building $X$ is a hyperbolic TDLC-group, whenever $\mathrm{Aut}(X)$ acts with finitely many orbits on $X$. Examples where this result applies include hyperbolic Bourdon's buildings.
We revisit the construction of small cancellation quotients of amalgamated free products, and verify that it provides examples of hyperbolic TDLC-groups of rational discrete cohomological dimension $2$ when applied to amalgamated products of profinite groups over open subgroups.
We raise the question of whether our main result can be extended to locally compact hyperbolic groups if rational discrete cohomological dimension is replaced by asymptotic dimension. We prove that this is the case for discrete groups and sketch an argument for TDLC-groups.
△ Less
Submitted 29 April, 2021; v1 submitted 21 August, 2019;
originally announced August 2019.
-
A Note on "Spaces of Topological Complexity One"
Authors:
Ramandeep Singh Arora
Abstract:
Here we give a reformulation of a key lemma in the paper [2], "Spaces of Topological Complexity One", which is necessary due to an oversight.
Here we give a reformulation of a key lemma in the paper [2], "Spaces of Topological Complexity One", which is necessary due to an oversight.
△ Less
Submitted 10 September, 2019; v1 submitted 14 April, 2019;
originally announced April 2019.
-
Subgroups of word hyperbolic groups in rational dimension 2
Authors:
Shivam Arora,
Eduardo Martínez-Pedroza
Abstract:
A result of Gersten states that if $G$ is a hyperbolic group with integral cohomological dimension $\mathsf{cd}_{\mathbb{Z}}(G)=2$ then every finitely presented subgroup is hyperbolic. We generalize this result for the rational case $\mathsf{cd}_{\mathbb{Q}}(G)=2$. In particular, our result applies to the class of torsion-free hyperbolic groups $G$ with $\mathsf{cd}_{\mathbb{Z}}(G)=3$ and…
▽ More
A result of Gersten states that if $G$ is a hyperbolic group with integral cohomological dimension $\mathsf{cd}_{\mathbb{Z}}(G)=2$ then every finitely presented subgroup is hyperbolic. We generalize this result for the rational case $\mathsf{cd}_{\mathbb{Q}}(G)=2$. In particular, our result applies to the class of torsion-free hyperbolic groups $G$ with $\mathsf{cd}_{\mathbb{Z}}(G)=3$ and $\mathsf{cd}_{\mathbb{Q}}(G)=2$ discovered by Bestvina and Mess.
△ Less
Submitted 17 December, 2020; v1 submitted 22 November, 2018;
originally announced November 2018.
-
Constructing Picard curves with complex multiplication using the Chinese Remainder Theorem
Authors:
Sonny Arora,
Kirsten Eisentraeger
Abstract:
We give a new algorithm for constructing Picard curves over a finite field with a given endomorphism ring. This has important applications in cryptography since curves of genus 3 allow for smaller key sizes than elliptic curves. For a sextic CM-field $K$ containing the cube roots of unity, we define and compute certain class polynomials modulo small primes and then use the Chinese Remainder Theore…
▽ More
We give a new algorithm for constructing Picard curves over a finite field with a given endomorphism ring. This has important applications in cryptography since curves of genus 3 allow for smaller key sizes than elliptic curves. For a sextic CM-field $K$ containing the cube roots of unity, we define and compute certain class polynomials modulo small primes and then use the Chinese Remainder Theorem to construct the class polynomials over the rationals. We also give some examples.
△ Less
Submitted 17 July, 2018; v1 submitted 1 March, 2018;
originally announced March 2018.
-
The twisting Sato-Tate group of the curve $y^2 = x^{8} - 14x^4 + 1$
Authors:
Sonny Arora,
Victoria Cantoral-Farfán,
Aaron Landesman,
Davide Lombardo,
Jackson S. Morrow
Abstract:
We determine the twisting Sato-Tate group of the genus $3$ hyperelliptic curve $y^2 = x^{8} - 14x^4 + 1$ and show that all possible subgroups of the twisting Sato-Tate group arise as the Sato-Tate group of an explicit twist of $y^2 = x^{8} - 14x^4 + 1$. Furthermore, we prove the generalized Sato-Tate conjecture for the Jacobians of all $\mathbb Q$-twists of the curve $y^2 = x^{8} - 14x^4 + 1$.
We determine the twisting Sato-Tate group of the genus $3$ hyperelliptic curve $y^2 = x^{8} - 14x^4 + 1$ and show that all possible subgroups of the twisting Sato-Tate group arise as the Sato-Tate group of an explicit twist of $y^2 = x^{8} - 14x^4 + 1$. Furthermore, we prove the generalized Sato-Tate conjecture for the Jacobians of all $\mathbb Q$-twists of the curve $y^2 = x^{8} - 14x^4 + 1$.
△ Less
Submitted 26 July, 2017; v1 submitted 24 August, 2016;
originally announced August 2016.
-
z-Classes in finite groups of conjugate type (n,1)
Authors:
Shivam Arora,
Krishnendu Gongopadhyay
Abstract:
Two elements in a group $G$ are said to $z$-equivalent or to be in the same $z$-class if their centralizers are conjugate in $G$. In \cite{kkj}, it was proved that a non-abelian $p$-group $G$ can have at most $\frac{p^k-1}{p-1} +1$ number of $z$-classes, where $|G/Z(G)|=p^k$. In this note, we characterize the $p$-groups of conjugate type $(n,1)$ attaining this maximal number. As a corollary, we ch…
▽ More
Two elements in a group $G$ are said to $z$-equivalent or to be in the same $z$-class if their centralizers are conjugate in $G$. In \cite{kkj}, it was proved that a non-abelian $p$-group $G$ can have at most $\frac{p^k-1}{p-1} +1$ number of $z$-classes, where $|G/Z(G)|=p^k$. In this note, we characterize the $p$-groups of conjugate type $(n,1)$ attaining this maximal number. As a corollary, we characterize $p$-groups having prime order commutator subgroup and maximal number of $z$-classes.
△ Less
Submitted 4 May, 2016;
originally announced May 2016.
-
Euclidean distortion and the Sparsest Cut
Authors:
Sanjeev Arora,
James R. Lee,
Assaf Naor
Abstract:
We prove that every $n$-point metric space of negative type (and, in particular, every $n$-point subset of $L_1$) embeds into a Euclidean space with distortion $O(\sqrt{\log n} \cdot\log \log n)$, a result which is tight up to the iterated logarithm factor. As a consequence, we obtain the best known polynomial-time approximation algorithm for the Sparsest Cut problem with general demands. Namely…
▽ More
We prove that every $n$-point metric space of negative type (and, in particular, every $n$-point subset of $L_1$) embeds into a Euclidean space with distortion $O(\sqrt{\log n} \cdot\log \log n)$, a result which is tight up to the iterated logarithm factor. As a consequence, we obtain the best known polynomial-time approximation algorithm for the Sparsest Cut problem with general demands. Namely, if the demand is supported on a subset of size $k$, we achieve an approximation ratio of $O(\sqrt{\log k}\cdot \log \log k)$.
△ Less
Submitted 8 August, 2005;
originally announced August 2005.
-
Learning mixtures of separated nonspherical Gaussians
Authors:
Sanjeev Arora,
Ravi Kannan
Abstract:
Mixtures of Gaussian (or normal) distributions arise in a variety of application areas. Many heuristics have been proposed for the task of finding the component Gaussians given samples from the mixture, such as the EM algorithm, a local-search heuristic from Dempster, Laird and Rubin [J. Roy. Statist. Soc. Ser. B 39 (1977) 1-38]. These do not provably run in polynomial time. We present the first…
▽ More
Mixtures of Gaussian (or normal) distributions arise in a variety of application areas. Many heuristics have been proposed for the task of finding the component Gaussians given samples from the mixture, such as the EM algorithm, a local-search heuristic from Dempster, Laird and Rubin [J. Roy. Statist. Soc. Ser. B 39 (1977) 1-38]. These do not provably run in polynomial time. We present the first algorithm that provably learns the component Gaussians in time that is polynomial in the dimension. The Gaussians may have arbitrary shape, but they must satisfy a ``separation condition'' which places a lower bound on the distance between the centers of any two component Gaussians. The mathematical results at the heart of our proof are ``distance concentration'' results--proved using isoperimetric inequalities--which establish bounds on the probability distribution of the distance between a pair of points generated according to the mixture. We also formalize the more general problem of max-likelihood fit of a Gaussian mixture to unstructured data.
△ Less
Submitted 22 March, 2005;
originally announced March 2005.
