• Corpus ID: 219708450

Learning Dynamics Models with Stable Invariant Sets

@inproceedings{Takeishi2021LearningDM,
  title={Learning Dynamics Models with Stable Invariant Sets},
  author={Naoya Takeishi and Y. Kawahara},
  booktitle={AAAI},
  year={2021}
}
Stable invariant sets are an essential notion in the analysis and application of dynamical systems. It is thus of great interest to learn dynamical systems with provable existence of stable invariant sets. However, existing methods can only deal with the stability of discrete equilibria, which hinders many applications. In this paper, we propose a method to ensure that a learned dynamics model has a stable invariant set of general classes. To this end, we modify a base dynamics model using a… 

Figures from this paper

Promoting global stability in data-driven models of quadratic nonlinear dynamics
TLDR
This work proposes that this theorem should be a standard diagnostic for the stability of projection-based and data-driven models, and illustrates how to modify the objective function in machine learning algorithms to promote globally stable models, with implications for the modeling of fluid and plasma flows.
Koopman Spectrum Nonlinear Regulator and Provably Efficient Online Learning
TLDR
This work presents a novel paradigm of controlling nonlinear systems via the minimization of the Koopman spectrum cost: a cost over the Koopsman operator of the controlled dynamics that induces a broader class of dynamical behaviors that evolve over stable manifolds such as nonlinear oscillators, closed loops, and smooth movements.
Symplectic Adjoint Method for Exact Gradient of Neural ODE with Minimal Memory
TLDR
The symplectic adjoint method is proposed, which obtains the exact gradient (up to rounding error) with a footprint proportional to the number of uses plus the network size and is robust to a rounding error among competitive methods.
Learning Deep Input-Output Stable Dynamics
TLDR
This study focuses on input-output stable systems, exhibiting robustness against unexpected stimuli and noise, and proposes a method to learn nonlinear systems guaranteeing the input- Output stability.
Learning Stabilizable Deep Dynamics Models
When neural networks are used to model dynamics, properties such as stability of the dynamics are generally not guaranteed. In contrast, there is a recent method for learning the dynamics of

References

SHOWING 1-10 OF 70 REFERENCES
Learning stable dynamical systems using contraction theory
This paper discusses the learning of robot point-to-point motions via non-linear dynamical systems and Gaussian Mixture Regression (GMR). The novelty of the proposed approach consists in guaranteeing
Learning Stable Stochastic Nonlinear Dynamical Systems
TLDR
This paper focuses on identifying a state-dependent coefficient form of the nonlinear stochastic model which is globally asymptotically stable according to probabilistic Lyapunov methods and compares the approach to other state of the art methods on real-world datasets in terms of flexibility and stability.
Neural learning of stable dynamical systems based on data-driven Lyapunov candidates
TLDR
A neural learning scheme that estimates stable dynamical systems from demonstrations based on a two-stage process: first, a data-driven Lyapunov function candidate is estimated, and stability is incorporated by means of a novel method to respect local constraints in the neural learning.
Learning Stable Deep Dynamics Models
TLDR
It is shown that such learning systems are able to model simple dynamical systems and can be combined with additional deep generative models to learn complex dynamics, such as video textures, in a fully end-to-end fashion.
Learning Stable Nonlinear Dynamical Systems With Gaussian Mixture Models
TLDR
A learning method is proposed, which is called Stable Estimator of Dynamical Systems (SEDS), to learn the parameters of the DS to ensure that all motions closely follow the demonstrations while ultimately reaching and stopping at the target.
Safe Model-based Reinforcement Learning with Stability Guarantees
TLDR
This paper presents a learning algorithm that explicitly considers safety, defined in terms of stability guarantees, and extends control-theoretic results on Lyapunov stability verification and shows how to use statistical models of the dynamics to obtain high-performance control policies with provable stability certificates.
Learning Dynamical Systems using Local Stability Priors
A coupled computational approach to simultaneously learn a vector field and the region of attraction of an equilibrium point from generated trajectories of the system is proposed. The nonlinear
...
...