30 March 2020 in Physical Review Research
Reservoir computing systems, a class of recurrent neural networks, have recently been exploited for model-free, data-based prediction of the state evolution of a variety of chaotic dynamical systems. The prediction horizon demonstrated has been about half dozen Lyapunov time. Is it possible to significantly extend the prediction time beyond what has been achieved so far? We articulate a scheme incorporating time-dependent but sparse data inputs into reservoir computing and demonstrate that such rare “updates” of the actual state practically enable an arbitrarily long prediction horizon for a variety of chaotic systems. A physical understanding based on the theory of temporal synchronization is developed.
28 January 2021 in Physical Review Research
To predict a critical transition due to parameter drift without relying on a model is an outstanding problem in nonlinear dynamics and applied fields. A closely related problem is to predict whether the system is already in or if the system will be in a transient state preceding its collapse. We develop a model-free, machine-learning-based solution to both problems by exploiting reservoir computing to incorporate a parameter input channel. We demonstrate that, when the machine is trained in the normal functioning regime with a chaotic attractor (i.e., before the critical transition), the transition point can be predicted accurately. Remarkably, for a parameter drift through the critical point, the machine with the input parameter channel is able to predict not only that the system will be in a transient state, but also the distribution of the transient lifetimes and their average before the final collapse, revealing an important physical property of transient chaos.
22 June 2021 in Physical Review Research
In realistic systems of coupled oscillators, it is desired to predict the onset of synchronization where the system equations are unknown, raising the need to develop a prediction framework that is model free and fully data driven. We show that this challenging problem can be addressed with machine learning. In particular, exploiting reservoir computing or echo state networks, we employ a “parameter-aware” scheme to train the neural machine using time series acquired from a small number of distinct asynchronous states in the parameter regime prior to the onset of synchronization. The trained machine can then be used to predict the synchronization transition through tuning the control parameter. We demonstrate the power of the machine learning-based framework using two types of synchronization behaviors: Complete synchronization in coupled identical chaotic oscillators and the phase synchronization in coupled nonidentical phase oscillators, which are representative of the collective dynamics in coupled systems. In addition, we design our numerical experiments such that two transition scenarios are covered: Smooth (second-order) and explosive (first-order) transitions that represent the generic types of phase transition in nonlinear physical systems. A remarkable feature is that, for the network systems exhibiting explosive (first-order) transition, the machine learning scheme is capable of predicting not only the locations of the transition points associated with the forward and backward transition paths but also the hysteresis between the two paths.
6 August 2021 in Physical Review E
Reconstructing the Kolmogorov-Arnold-Moser (KAM) dynamics diagram of Hamiltonian system from the time series of a limited number of parameters is an outstanding question in nonlinear science, especially when the Hamiltonian governing the system dynamics is unknown. Here we demonstrate that this question can be addressed by the machine learning approach knowing as reservoir computing (RC). Specifically, we show that without prior knowledge about the Hamilton equations of motion, the trained RC is able to not only predict the short-term evolution of the system state, but also replicate the long-term ergodic properties of the system dynamics. Furthermore, using the architecture of parameter-aware RC, we show that the RC trained by the time series acquired at a handful parameters is able to reconstruct the entire KAM dynamics diagram with a high precision by tuning a control parameter externally. The feasibility and efficiency of the learning techniques are demonstrated in two classical nonlinear Hamiltonian systems, namely, the double-pendulum oscillator and the standard map. Our study indicates that, as a complex dynamical system, RC is able to learn from data the Hamiltonian.
2 July 2021 in Journal of Physics: Complexity
An emerging paradigm for predicting the state evolution of chaotic systems is machine learning with reservoir computing, the core of which is a dynamical network of artificial neurons. Through training with measured time series, a reservoir machine can be harnessed to replicate the evolution of the target chaotic system for some amount of time, typically about half dozen Lyapunov times. Recently, we developed a reservoir computing framework with an additional parameter channel for predicting system collapse and chaotic transients associated with crisis. It was found that the crisis point after which transient chaos emerges can be accurately predicted. The idea of adding a parameter channel to reservoir computing has also been used by others to predict bifurcation points and distinct asymptotic behaviors. In this paper, we address three issues associated with machine-generated transient chaos. First, we report the results from a detailed study of the statistical behaviors of transient chaos generated by our parameter-aware reservoir computing machine. When multiple time series from a small number of distinct values of the bifurcation parameter, all in the regime of attracting chaos, are deployed to train the reservoir machine, it can generate the correct dynamical behavior in the regime of transient chaos of the target system in the sense that the basic statistical features of the machine generated transient chaos agree with those of the real system. Second, we demonstrate that our machine learning framework can reproduce intermittency of the target system. Third, we consider a system for which the known methods of sparse optimization fail to predict crisis and demonstrate that our reservoir computing scheme can solve this problem. These findings have potential applications in anticipating system collapse as induced by, e.g., a parameter drift that places the system in a transient regime.
18 April 2016 in Physical Review E
Although the set of permutation symmetries of a complex network could be very large, few of them give rise to stable synchronous patterns. Here we present a general framework and develop techniques for controlling synchronization patterns in complex network of coupled chaotic oscillators. Specifically, according to the network permutation symmetry, we design a small-size and weighted network, namely the control network, and use it to control the large-size complex network by means of pinning coupling. We argue mathematically that for any of the network symmetries, there always exists a critical pinning strength beyond which the unstable synchronous pattern associated to this symmetry can be stabilized. The feasibility of the control method is verified by numerical simulations of both artificial and real-world networks and demonstrated experimentally in systems of coupled chaotic circuits. Our studies show the controllability of synchronous patterns in complex networks of coupled chaotic oscillators.