Learning the tangent space of dynamical instabilities from data.

Chaos

Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA.

Published: November 2019

For a large class of dynamical systems, the optimally time-dependent (OTD) modes, a set of deformable orthonormal tangent vectors that track directions of instabilities along any trajectory, are known to depend "pointwise" on the state of the system on the attractor but not on the history of the trajectory. We leverage the power of neural networks to learn this "pointwise" mapping from the phase space to OTD space directly from data. The result of the learning process is a cartography of directions associated with strongest instabilities in the phase space. Implications for data-driven prediction and control of dynamical instabilities are discussed.

Download full-text PDF

Source
http://dx.doi.org/10.1063/1.5120830DOI Listing

Publication Analysis

Top Keywords

dynamical instabilities
8
phase space
8
learning tangent
4
space
4
tangent space
4
space dynamical
4
instabilities
4
instabilities data
4
data large
4
large class
4

Similar Publications

The applications of resveratrol (RES) and puerarin (PUE) with notable physiological functions are greatly limited in functional food and pharmaceutical industries due to their poor water solubility and chemical instability. Accordingly, co-loading of RES and PUE into chitosan-based nanoparticles (NPs) is performed here by an anti-solvent method to improve their bioavailability. The fabricated NPs at 8:1 mass ratio of carboxymethyl chitosan (CMC) to chitosan hydrochloride (CHC) with the particle size of 375.

View Article and Find Full Text PDF

This study offers a comprehensive analysis of the Perturbed Schrödinger -Hirota Equation (PSHE), crucial for understanding soliton dynamics in modern optical communication systems. We extended the traditional Nonlinear Schrödinger Equation (NLSE) to include higher-order nonlinearities and spatiotemporal dispersion, capturing the complexities of light pulse propagation. Employing the modified auxiliary equation method and Adomian Decomposition Method (ADM), we derived a spectrum of exact traveling wave solutions, encompassing exponential, rational, trigonometric, and hyperbolic functions.

View Article and Find Full Text PDF

Phocaeicola vulgatus shapes the long-term growth dynamics and evolutionary adaptations of Clostridioides difficile.

Cell Host Microbe

December 2024

Department of Biochemistry, University of Wisconsin-Madison, Madison, WI, USA; Department of Chemical & Biological Engineering, University of Wisconsin-Madison, Madison, WI, USA; Department of Biomedical Engineering, Duke University, Durham, NC, USA; Department of Bacteriology, University of Wisconsin-Madison, Madison, WI, USA; Department of Biomedical Engineering, University of Wisconsin-Madison, Madison, WI, USA. Electronic address:

Clostridioides difficile can transiently or persistently colonize the human gut, posing a risk for infections. This colonization is influenced by complex molecular and ecological interactions with the human gut microbiota. By investigating C.

View Article and Find Full Text PDF

In this paper, a deep reinforcement learning (DRL) approach based on generative adversarial imitation learning (GAIL) and long short-term memory (LSTM) is proposed to resolve tracking control problems for robotic manipulators with saturation constraints and random disturbances, without learning the dynamic and kinematic model of the manipulator. Specifically, it limits the torque and joint angle to a certain range. Firstly, in order to cope with the instability problem during training and obtain a stability policy, soft actor-critic (SAC) and LSTM are combined.

View Article and Find Full Text PDF

Objective: Lateral ankle sprains (LASs) are prevalent in soccer and can affect long-term performance, injury recurrence and risk for chronic ankle instability. This case-control study examined functional impairments associated with LAS in professional soccer players aged 17-21.

Methods: 40 players were divided into 2 groups: 21 with a history of LAS and 19 healthy matched controls.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!