arXiv subjects

Jaesung Choi

Publications and source records attributed to Jaesung Choi.

7 recordsLinked to original sources

arXiv2026

Learning a quantitative criterion for distinguishing chaos from noise

Distinguishing chaos from noise using time-series data is fundamentally challenging because both exhibit irregular fluctuations and share many statistical and dynamical characteristics. Existing methods face two key limitations: temporally correlated noise can yield spurious signatures of chaos, and analyses of scalar time series often require explicit choices of embedding parameters. Here, we propose a purely data-driven method for distinguishing chaos and noise based on a reservoir-computing framework with a cross-prediction scheme. In the proposed approach, the model is trained to predict the future change of a variable from its current value, thereby combining a short-term predictability test with a test of the smoothness of deterministic flows. The recurrent structure of reservoir computing enables effective prediction of high-dimensional chaotic dynamics even from scalar time series without explicit delay-coordinate reconstruction, while the cross-prediction framework strongly suppresses spurious predictive correlations arising from noise. We apply the proposed method to diverse synthetic and empirical time series. Chaotic systems consistently yield strong correlations between the true and predicted future changes, whereas noise processes remain clearly separated in a low-correlation regime. The method also exhibits substantial robustness to practical limitations in empirical data, including measurement noise, limited data length, and increasing prediction lag. These results demonstrate that the squared Pearson correlation coefficient provides a simple quantitative criterion for distinguishing chaos from noise directly from observed time-series data.

nlin.CD↗

arXiv2026

Extrapolating the emergence of Hamiltonian chaos with random-feature Hamiltonian neural networks

Machine learning of Hamiltonian dynamics has driven growing interest in Hamiltonian neural networks (HNNs), which encode Hamilton's equations of motion into the learning architecture. Despite this progress, it remains unknown whether such networks can predict dynamical regimes absent from their training data, in particular the broad chaotic sea that emerges beyond the observed parameter interval. We address this question using a parameter-aware random-feature Hamiltonian neural network (RF-HNN). Trained using data from only a small number of control-parameter values at which invariant tori dominate, the RF-HNN predicts autonomous long-time dynamics at unseen parameter values where mixed phase space develops and chaotic regions expand, with no data from that regime used in training or model selection. The method is demonstrated across four two-degree-of-freedom Hamiltonian families, including the H\'enon-Heiles system. Using Poincar\'e-section geometry and finite-time Lyapunov exponents, we show that the RF-HNN reproduces the breakup of regular structures and the emergence and growth of chaotic regions, whereas conventionally trained HNNs with the same Hamiltonian structure remain too regular. These results show that what decides parameter extrapolation is not Hamiltonian structure alone but how the fitted Hamiltonian continues in the control parameter. To our knowledge, this is the first demonstration that a learned Hamiltonian can qualitatively extrapolate from predominantly regular dynamics into a broad chaotic sea absent from training.

nlin.CD↗

arXiv2025

Human brain state classification via permutation entropy of EEG phase dynamics across consciousness levels and inattentive-type ADHD

We analyze electroencephalography (EEG) signals using the ordinal pattern framework to investigate whether different human brain states can be distinguished based on the disorder of EEG dynamics. Rather than analyzing raw EEG signals, we focus on the principal mode of EEG phase dynamics, reflecting anterior-posterior information flow, and quantify disorder using permutation entropy. We apply this to two datasets: (i) EEG recordings from a general anesthesia protocol, and (ii) EEG recordings acquired in the resting state from healthy control subjects and individuals with inattentive-type attention deficit hyperactivity disorder (ADHD), including eyes-open and eyes-closed conditions. We find that the permutation entropy distributions exhibit a clear dependence on brain state. In particular, conscious, inattentive-type ADHD, and eyes closed conditions show lower mean values and larger standard deviations of permutation entropy. To evaluate the discriminative power of permutation entropy, we train classification models using permutation entropy as an input feature. The results show that the distinction between conscious and unconscious states can be reliably captured in the general-anesthesia dataset. In the resting-state dataset, eyes-open and eyes-closed conditions are distinguishable, whereas classification between control and inattentive-type ADHD groups does not show clear separability. This indicates that information not captured in ordinal patterns, such as the original time-series values, may play a more crucial role in detecting inattentive-type ADHD. Our findings demonstrate that permutation entropy derived from EEG phase dynamics provides an effective indicator of brain states, particularly in relation to consciousness, while also highlighting its limitations for identifying individuals with inattentive-type ADHD.

q-bio.NC↗

arXiv2024

Unsupervised Reservoir Computing for Multivariate Denoising of Severely Contaminated Signals

The interdependence and high dimensionality of multivariate signals present significant challenges for denoising, as conventional univariate methods often struggle to capture the complex interactions between variables. A successful approach must consider not only the multivariate dependencies of the desired signal but also the multivariate dependencies of the interfering noise. In our previous research, we introduced a method using machine learning to extract the maximum portion of ``predictable information" from univariate signal. We extend this approach to multivariate signals, with the key idea being to properly incorporate the interdependencies of the noise back into the interdependent reconstruction of the signal. The method works successfully for various multivariate signals, including chaotic signals and highly oscillating sinusoidal signals which are corrupted by spatially correlated intensive noise. It consistently outperforms other existing multivariate denoising methods across a wide range of scenarios.

cs.LG↗

arXiv2024

Signal-noise separation using unsupervised reservoir computing

Removing noise from a signal without knowing the characteristics of the noise is a challenging task. This paper introduces a signal-noise separation method based on time series prediction. We use Reservoir Computing (RC) to extract the maximum portion of "predictable information" from a given signal. Reproducing the deterministic component of the signal using RC, we estimate the noise distribution from the difference between the original signal and reconstructed one. The method is based on a machine learning approach and requires no prior knowledge of either the deterministic signal or the noise distribution. It provides a way to identify additivity/multiplicativity of noise and to estimate the signal-to-noise ratio (SNR) indirectly. The method works successfully for combinations of various signal and noise, including chaotic signal and highly oscillating sinusoidal signal which are corrupted by non-Gaussian additive/ multiplicative noise. The separation performances are robust and notably outstanding for signals with strong noise, even for those with negative SNR.

cs.LG↗

arXiv2019

Reservoir Computing based on Quenched Chaos

Reservoir computing(RC) is a brain-inspired computing framework that employs a transient dynamical system whose reaction to an input signal is transformed to a target output. One of the central problems in RC is to find a reliable reservoir with a large criticality, since computing performance of a reservoir is maximized near the phase transition. In this work, we propose a continuous reservoir that utilizes transient dynamics of coupled chaotic oscillators in a critical regime where sudden amplitude death occurs. This "explosive death" not only brings the system a large criticality which provides a variety of orbits for computing, but also stabilizes them which otherwise diverge soon in chaotic units. The proposed framework shows better results in tasks for signal reconstructions than RC based on explosive synchronization of regular phase oscillators. We also show that the information capacity of the reservoirs can be used as a predictive measure for computational capability of a reservoir at a critical point.

nlin.CD↗

arXiv2018

Critical Neuromorphic Computing based on Explosive Synchronization

Synchronous oscillations in neuronal ensembles have been proposed to provide a neural basis for the information processes in the brain. In this work, we present a neuromorphic computing algorithm based on oscillator synchronization in a critical regime. The algorithm uses the high dimensional transient dynamics perturbed by an input and translates it into proper output stream. One of the benefits of adopting coupled phase oscillators as neuromorphic elements is that the synchrony among oscillators can be finely tuned at a critical state. Especially near a critical state, the marginally synchronized oscillators operate with high efficiency and maintain better computing performances. We also show that explosive synchronization which is induced from specific neuronal connectivity produces more improved and stable outputs. This work provides a systematic way to encode computing in a large size coupled oscillators, which may be useful in designing neuromorphic devices.

cs.NE↗