Qin Lu

University of Georgia · United States

Unclaimed · AI-structured profileOpenAlex

Qin Lu (University of Georgia): Bayesian uncertainty in graph learning and underwater acoustic inversion

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-15) 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—

Aggregate output metrics for this OpenAlex identity are not yet available; only 2 works in the recent three-year window are visible. This axis would measure total publication volume, productivity pace, and the share of author-list positions, so it cannot be scored honestly at present.

Impact—

Impact would track citations, h-index, and i10-index over time, including per-year citation history. The OpenAlex record currently reports h-index 0 and i10-index 0 with no citation history, which is insufficient to characterize influence, so no score is given.

Focus0/100

The focus score is 0/100, reflecting a profile split exactly evenly, 0.5 versus 0.5, between two distinct lines: graph representation learning with uncertainty quantification, and Bayesian optimization for underwater acoustics. OpenAlex topics reinforce this breadth, spanning Gaussian processes and Bayesian inference (2 works) plus single works in graph neural networks, adversarial robustness, underwater acoustics, and bandit algorithms.

Collaboration36/100

The collaboration score is 36/100, based on a modest OpenAlex footprint of 5 distinct co-authors, matching the recurring collaborators in Evolve's data: Jinwen Xu, Gonzalo Mateos Buckstein, Heng Zhang, Haotian Xiang, and Florian Meyer. Both cataloged 2026 papers are multi-author arXiv works with Lu in the last author position, suggesting small specialized teams rather than large consortia.

Activity32/100

Activity scores 32/100: the OpenAlex record shows only 2 works in the recent three-year window, both appearing in 2026, and no counts-by-year history before that. This is more consistent with an early or newly indexed publishing profile than with demonstrated low engagement — the two 2026 arXiv papers arrived together.

Reproducibility55/100

Reproducibility scores 55/100, a mean across the 2 cataloged papers, both rated medium readiness. Evolve's review notes the Bayesian optimization paper ships no tests or code, and the online node classification paper leaves evidence gaps despite winning accuracy and NLL on all five benchmarks, so methodological detail is present but artifact availability is the main gap.

Research summary

Qin Lu is a machine-learning researcher at the University of Georgia whose bibliometric footprint, resolved from OpenAlex, splits evenly across two areas: graph representation learning with uncertainty quantification, and Bayesian optimization applied to underwater acoustics. Both papers currently in Evolve's catalog appeared on arXiv in 2026, with Lu in the last author position, a role often associated with senior or supervising contributions, although neither paper lists Lu as corresponding author. The first paper develops Bayesian optimization with kernel-ensemble Gaussian processes and disagreement-based acquisition for source localization and geoacoustic inversion, and Evolve's review finds it claims the best Bayesian-optimization results on the SWellEx-96 scenario, with solid evidence overall but a thin margin on real data and no released tests or code. The second extends Bayesian last layers to online node classification on inductive graphs under distribution shift, winning accuracy and NLL on all five benchmarks while leaving evidence gaps. OpenAlex topic data corroborates this spread: Gaussian processes and Bayesian inference (2 works) alongside single works in graph neural networks, adversarial robustness, underwater acoustics, and bandit algorithms. Traditional citation metrics (h-index 0, i10-index 0) and total output are not yet available, so this remains an early-stage profile.

Collaboration pattern

Qin Lu's recorded collaboration footprint is compact: 5 distinct co-authors in OpenAlex, all recurring in Evolve's data — Jinwen Xu, Gonzalo Mateos Buckstein, Heng Zhang, Haotian Xiang, and Florian Meyer. Both 2026 cataloged papers are multi-author arXiv works with Lu in the last author position, not corresponding author, a pattern consistent with senior or supervising involvement in small specialized teams spanning Bayesian methods, graph learning, and underwater acoustics.

Research keywords

Bayesian optimizationGaussian processesKernel ensembleAcquisition functionsGeoacoustic inversionSource localizationGraph neural networksBayesian last layerVariational inferenceOnline learningDistribution shiftNode classification

Highlights

Kernel-ensemble Bayesian optimization for underwater acoustic inversion
Evolve's review finds the kernel-ensemble Gaussian process approach with disagreement-based acquisition claims the best Bayesian-optimization results for SWellEx-96 source localization and geoacoustic inversion, with solid evidence overall but a thin margin on real data and no released tests or code.
Bayesian last layers for online node classification under distribution shift
The paper extends Bayesian last layers to online node classification on inductive graphs under distribution shift, winning accuracy and NLL on all five benchmarks, though Evolve's review notes remaining evidence gaps.

Timeline

2026First cataloged work appears on arXiv: Bayesian optimization with kernel ensembles and disagreement-based acquisition for source localization and acoustic inversion, with Qin Lu in the last author position.
2026Second arXiv paper appears: Online Bayesian Node Classification on Inductive Graphs under Distribution Shift, again with Qin Lu as last author and not corresponding author.
2026OpenAlex records 2 works for the year, involving 5 distinct co-authors including Jinwen Xu, Gonzalo Mateos Buckstein, Heng Zhang, Haotian Xiang, and Florian Meyer.
2026Both papers enter the Evolve catalog with medium reproducibility readiness, yielding a reproducibility mean of 55.

Papers in the Evolve catalog

Frequent co-authors

Jinwen Xu×1 Gonzalo Mateos Buckstein×1 Heng Zhang×1 Haotian Xiang×1 Florian Meyer×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 resolved from OpenAlex bibliographic data only; no ORCID is linked, and Qin Lu is a common name, so the profile may merge or split homonymous researchers.
Citation-based metrics (h-index 0, i10-index 0) and total publication counts are unavailable or empty in OpenAlex, so the output and impact axes are left unscored.
The Evolve catalog covers only 2 arXiv preprints from 2026; peer-reviewed venues, awards, and career details are outside the data scope.
Reproducibility scores come from Evolve's manual review of those 2 papers (mean 55) and may not generalize to work not yet in the catalog.