English
Related papers

Related papers: Backpropagation in hyperbolic chaos via adjoint sh…

200 papers

The classical spectral theorem completely describes self-adjoint operators on finite dimensional inner product vector spaces as linear combinations of orthogonal projections onto pairwise orthogonal subspaces. We prove a similar theorem for…

Rings and Algebras · Mathematics 2017-10-03 Camilo Sanabria Malagón

We outline a global approach to scattering theory in one dimension that allows for the description of a large class of scattering systems and their $\mathcal{P}$-, $\mathcal{T}$-, and $\mathcal{P}\mathcal{T}$-symmetries. In particular, we…

Quantum Physics · Physics 2019-01-25 Ali Mostafazadeh

Entrainment of limit cycles by chaos [1] is discovered numerically through specially designed unidirectional coupling of two glow discharge-semiconductor systems. By utilizing the auxiliary system approach [2], it is verified that the…

Chaotic Dynamics · Physics 2014-06-18 Marat Akhmet , Ismail Rafatov , Mehmet Onur Fen

This paper deals mainly with some aspects of the adjointable operators on Hilbert $C^*$-modules. A new tool called the generalized polar decomposition for each adjointable operator is introduced and clarified. As an application, the general…

Functional Analysis · Mathematics 2024-04-25 Xiaofeng Zhang , Xiaoyi Tian , Qingxiang Xu

Semi-Lagrangian methods have traditionally been developed in the framework of hyperbolic equations, but several extensions of the Semi-Lagrangian approach to diffusion and advection--diffusion problems have been proposed recently. These…

Numerical Analysis · Mathematics 2014-05-20 L. Bonaventura , R. Ferretti

This paper addresses inverse problems (in a broad sense) for two classes of multivariate neural network (NN) operators, with particular emphasis on saturation results, and both analytical and semi-analytical inverse theorems. One of the key…

Functional Analysis · Mathematics 2025-05-13 Danilo Costarelli

A key step in reverse engineering neural networks is to decompose them into simpler parts that can be studied in relative isolation. Linear parameter decomposition -- a framework that has been proposed to resolve several issues with current…

Machine Learning · Computer Science 2025-09-05 Lucius Bushnaq , Dan Braun , Lee Sharkey

We introduce Equilibrium Propagation, a learning framework for energy-based models. It involves only one kind of neural computation, performed in both the first phase (when the prediction is made) and the second phase of training (after the…

Machine Learning · Computer Science 2017-03-30 Benjamin Scellier , Yoshua Bengio

We study a class of slow-fast Hamiltonian systems with any finite number of degrees of freedom, but with at least one slow one and two fast ones. At $% \epsilon =0$ the slow dynamics is frozen. We assume that the frozen system (i.e. the…

Dynamical Systems · Mathematics 2015-05-13 Niklas Brännström , Emiliano De Simone , Vassili Gelfreich

We apply spatial dynamical-systems techniques to prove that certain spatiotemporal patterns in reversible reaction-diffusion equations undergo snaking bifurcations. That is, in a narrow region of parameter space, countably many branches of…

Dynamical Systems · Mathematics 2025-07-23 Timothy Roberts , Bjorn Sandstede

We introduce quantum circuits in two and three spatial dimensions which are classically simulable, despite producing a high degree of operator entanglement. We provide a partial characterization of these "automaton" quantum circuits, and…

Statistical Mechanics · Physics 2019-12-10 Jason Iaconis , Sagar Vijay , Rahul Nandkishore

We investigate superdiffusion for stochastic processes generated by nonuniformly hyperbolic system models, in terms of the convergence of rescaled distributions to the normal distribution following the abnormal central limit theorem, which…

Dynamical Systems · Mathematics 2017-09-05 Luke Mohr , Hong-Kun Zhang

This paper is devoted to the definition and analysis of the spectral shift function (SSF) associated with non-self-adjoint perturbations of self-adjoint operators. Motivated by applications in scattering theory, we consider both trace-class…

Mathematical Physics · Physics 2026-03-24 Vincent Bruneau , Nicolas Frantz , François Nicoleau

We propose to build a combinatorial invariant, called the spectral monodromy, from the spectrum of a single non-selfadjoint h-pseudodifferential operator with two degrees of freedom in the semi-classical limit. Our inspiration comes from…

Mathematical Physics · Physics 2015-06-15 Quang Sang Phan

We present a parallel data-driven strategy to identify finite-dimensional functional spaces invariant under the Koopman operator associated to an unknown dynamical system. We build on the Symmetric Subspace Decomposition (SSD) algorithm, a…

Systems and Control · Electrical Eng. & Systems 2021-03-30 Masih Haseli , Jorge Cortés

We consider linear, time-dependent and skew-adjoint perturbations of periodic transport equations on the one-dimensional torus. We describe the long-time behavior of solutions for all non-degenerate perturbations in resonant regime, proving…

Analysis of PDEs · Mathematics 2025-11-25 Maria Teresa Rotolo

This paper proposes an algorithmic framework for solving parametric optimization problems which we call adjoint-based predictor-corrector sequential convex programming. After presenting the algorithm, we prove a contraction estimate that…

Optimization and Control · Mathematics 2011-09-14 Q. Tran Dinh , C. Savorgnan , M. Diehl

We present an extension of the chaos-assisted tunneling mechanism to spatially periodic lattice systems. We demonstrate that driving such lattice systems in an intermediate regime of modulation maps them onto tight-binding Hamiltonians with…

The abstract theory of self-adjoint extensions of symmetric operators is used to construct self-adjoint realizations of a second-order elliptic operator on $\mathbb{R}^{n}$ with linear boundary conditions on (a relatively open part of) a…

Analysis of PDEs · Mathematics 2016-04-12 A. Mantile , A. Posilicano , M. Sini

We study stochastic gradient descent (SGD) for composite optimization problems with $N$ sequential operators subject to perturbations in both the forward and backward passes. Unlike classical analyses that treat gradient noise as additive…

Optimization and Control · Mathematics 2026-02-25 Boao Kong , Hengrui Zhang , Kun Yuan
‹ Prev 1 8 9 10 Next ›