Random Dynamical Systems and Attractors

Random dynamical systems (RDS) constitute a mathematical framework for analysing the evolution of systems subject to stochastic influences. In such systems, noise enters either additively or multiplicatively, leading to trajectories that depend on both initial conditions and realisations of the random input. Central to the theory of RDS is the concept of attractors, invariant sets towards which the system converges over long time horizons. Unlike deterministic attractors, random attractors capture the interplay between deterministic dynamics and random perturbations, providing a statistical description of long-term behaviour. Key properties include measurability with respect to the underlying probability space, invariance under the stochastic flow, and attraction of all bounded sets in an appropriate phase space. Research in this field spans diverse equations—from reaction-diffusion and wave equations to fluid dynamics models—on both bounded and unbounded domains. Recent advances have emphasised the existence, uniqueness and robustness of attractors under parameter variations, as well as the fine structure of their fractal dimensions and continuity properties under noise modulation. This body of work underpins applications ranging from climate modelling to neural network dynamics and materials science, where stochastic forcing is intrinsic to system behaviour.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Recent analyses of impulsive generalised semiflows have extended traditional attractor theory to systems with state-dependent jumps. New criteria establish the existence of global attractors even when solution uniqueness fails, and demonstrate upper-semicontinuity of attractors under perturbations. These results encompass both ordinary and partial differential equations subject to instantaneous impulses, offering a unified view of long-term behaviour in hybrid dynamical contexts.

Investigations into fractional nonclassical diffusion equations driven by coloured noise have illustrated the existence and uniqueness of pullback random attractors in fractional Sobolev spaces. Employing uniform tail estimates and spectral decomposition, these studies overcome the non-compactness of embeddings on unbounded domains. Moreover, they prove upper-semicontinuity of attractors as the correlation time of the noise vanishes, bridging the gap between coloured and white noise regimes.

Work on stochastic FitzHugh–Nagumo systems with coloured noise has demonstrated the pathwise existence and uniqueness of random attractors for both linear and nonlinear diffusion. Detailed comparisons reveal that attractors under multiplicative coloured noise converge to those for white noise as noise colour parameters tend to zero. This sheds light on the robustness of neural signal models to the temporal structure of environmental fluctuations.

Random Dynamical Systems and Attractors publication trend

The graph below shows the total number of articles in random dynamical systems and attractors across all publications each year (not limited to Nature Index journals).

Technical terms

Random dynamical system: A system of differential or difference equations whose evolution law depends on stochastic processes, formalised via a cocycle property on a probability space.

Global attractor: A compact invariant set that attracts every bounded subset of the phase space under the flow of the system, describing its long-term deterministic or random behaviour.

Pullback attractor: A family of sets parameterised by time that attracts all past states when pulled back along the flow, especially suited to non-autonomous or random settings.

Asymptotic compactness: A property whereby the images of bounded sets under the flow eventually become pre-compact, ensuring the existence of attractors in infinite-dimensional spaces.

Upper-semicontinuity: A continuity property of attractors under parameter changes, meaning that limiting attractors lie inside any given neighbourhood of the perturbed attractors for small perturbations.

References

1. Long-time behavior for impulsive generalized semiflows. Nonlinear Analysis Hybrid Systems (2024).
2. Random dynamics of fractional nonclassical diffusion equations driven by colored noise. Discrete and Continuous Dynamical Systems (2019).
3. Asymptotic behavior of random fitzhugh-nagumo systems driven by colored noise. Discrete and Continuous Dynamical Systems - B (2018).