Xiaole Zhang

—

Unclaimed · AI-structured profileORCID OpenAlex

First-author theory work on learning fractional-order dynamics from a single trajectory with non-asymptotic guarantees

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

—

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-09-20) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.

Research profile

Output—

Insufficient data: no publication counts or yearly output records are available yet. This axis would measure research productivity — how many papers appear, where, and how often — and is left unscored rather than estimated.

Impact—

Insufficient data: no citation counts, h-index, or venue-level influence metrics are provided. This axis would measure the uptake of the work by other researchers — citations, downloads, and scholarly attention — and is left unscored.

Focus100/100

A perfect focus score of 100 follows from the field distribution: all reviewed activity sits in one field — system identification and control theory with statistical learning for dynamical systems. The single reviewed 2026 paper concentrates tightly on fractional-order system identification, and the keyword set (fractional-order systems, long-range dependence, single-trajectory learning) shows a coherent, undispersed research agenda, though it rests on one paper.

Collaboration39/100

The collaboration score of 39 reflects a compact network: six distinct co-authors appear in the data, anchored by a recurring group that includes Ziyi Zhang, Zehao Zhao, Stephen Tu, Guannan Qu, Yorie Nakahira, and Paul Bogdan. The reviewed 2026 paper is first-authored but not corresponding, suggesting a contributing role within a small, theory-oriented team rather than a broad, distributed collaboration web.

Activity—

Insufficient data: the snapshot covers one 2026 arXiv preprint, with no multi-year publication rhythm to assess. This axis would measure ongoing research tempo — recent papers, preprints, and sustained engagement — and is left unscored.

Reproducibility55/100

Reproducibility scores 55 (reproMean over 1 evaluated paper). The reviewed FO-GS paper is rated medium reproducibility readiness: the grid-search OLS estimator is concretely specified and backed by synthetic experiments, but Evolve's review flags the EEG real-data protocol as having notable statistical weaknesses, which tempers confidence in independently reproducing the real-world results without additional detail.

Research summary

This is an early-stage profile of a researcher working at the intersection of system identification, control theory, and statistical learning for dynamical systems. The Evolve catalog currently holds one reviewed paper, a 2026 arXiv preprint authored in the first position: 'Learning Fractional-Order Dynamics from a Single Trajectory'. The paper proposes FO-GS, a row-wise grid-search OLS estimator that learns fractional-order linear time-invariant systems from a single observed trajectory — a setting motivated by non-Markovian dynamics and long-range dependence. Its central theoretical contribution is the first non-asymptotic O(t^-1/2) guarantee for jointly recovering the fractional order and the system matrix, demonstrated on both synthetic data and EEG recordings. The bibliometric snapshot is deliberately narrow: the field distribution contains exactly one field, the collaboration score of 39 reflects six distinct co-authors (including Ziyi Zhang, Zehao Zhao, Stephen Tu, Guannan Qu, Yorie Nakahira, and Paul Bogdan), and reproducibility averages 55 across one evaluated paper at medium readiness. Evolve's review praises the theory while flagging statistical weaknesses in the EEG protocol. Output, impact, and activity metrics are not yet available and are left unscored. This page will deepen as further papers, citations, and affiliations enter the catalog; identity is resolved via ORCID 0000-0002-8349-8101 from public bibliographic data.

Collaboration pattern

The co-author footprint is compact: six distinct collaborators appear in the data, and the recurring set includes Ziyi Zhang, Zehao Zhao, Stephen Tu, Guannan Qu, Yorie Nakahira, and Paul Bogdan. On the reviewed 2026 paper Zhang is first author but not corresponding, consistent with a hands-on technical contributor inside a small, theory-driven group rather than an independent lab leading large collaborations.

Research keywords

fractional-order systemssystem identificationGrunwald-Letnikov differencenon-Markovian dynamicslong-range dependencesingle-trajectory learninglinear time-invariant (LTI) systemsnon-asymptotic estimation guaranteesordinary least squares (OLS)grid-search estimationEEG signal analysiscontrol theory

Highlights

FO-GS: learning fractional-order dynamics from one trajectory
First-author 2026 arXiv preprint proposing a row-wise grid-search OLS estimator that recovers fractional-order LTI systems from a single observed trajectory, targeting non-Markovian dynamics with long-range dependence via the Grunwald-Letnikov difference.
First non-asymptotic O(t^-1/2) guarantees for joint recovery
Evolve's verdict credits the paper with the first non-asymptotic O(t^-1/2) guarantees for jointly recovering the fractional order and the system matrix — the core theory-first contribution of the work.
Validation spans synthetic data and EEG, with caveats
The estimator is demonstrated on synthetic systems and real EEG recordings; Evolve's review notes the real-world protocol has notable statistical weaknesses, and reproducibility readiness is rated medium (reproMean 55 from one paper).
A tightly focused agenda at the catalog level
All reviewed activity falls in a single catalog field — system identification and control theory / statistical learning for dynamical systems — producing a focus score of 100, with keywords clustering on fractional-order models and single-trajectory learning.

Timeline

2026Posted the first-author arXiv preprint 'Learning Fractional-Order Dynamics from a Single Trajectory', entering Evolve's catalog under system identification and control theory / statistical learning for dynamical systems.
2026Proposed FO-GS, a row-wise grid-search OLS estimator identifying fractional-order LTI systems from a single trajectory.
2026Established the first non-asymptotic O(t^-1/2) guarantees for jointly recovering the fractional order and the system matrix, supported by synthetic experiments.
2026Demonstrated the estimator on real EEG recordings, extending toward non-Markovian neural signals; Evolve's review flags statistical weaknesses in this protocol.

Papers in the Evolve catalog

2026-09-17first author
Learning Fractional-Order Dynamics from a Single Trajectory

Frequent co-authors

Ziyi Zhang×1 Zehao Zhao×1 Stephen Tu×1 Guannan Qu×1 Yorie Nakahira×1 Paul Bogdan×1

About this profile

This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.

Identity is name-resolved from public bibliographic data via ORCID 0000-0002-8349-8101; no institutional affiliation or country is present in the data.
Output, impact, and activity dimensions are unscored: no citation counts, publication volumes, or multi-year records were available.
Field distribution, the focus score (100), and the reproducibility mean (55) rest on a single reviewed paper; scores may change as the catalog grows.
The 2026 paper is an arXiv preprint; peer-review venue and acceptance status are not captured in the data.