Justin Kay

Massachusetts Institute of Technology · United States

Unclaimed · AI-structured profileOpenAlex

Building ML benchmarks and active-learning tools for wildlife movement forecasting and biodiversity surveys (MIT)

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-08) 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—

Not yet scored. The OpenAlex snapshot carries only early signals (h-index 3 across 14 indexed works), too thin to aggregate reliably. This axis would measure publication volume, venue quality, and author role across the record.

Impact—

Not yet scored. Citation uptake is too early-stage (h-index 3, i10-index 2) to support a stable estimate. This axis would measure citation counts, downstream reuse, and reach across both ecology and machine-learning audiences.

Focus0/100

Scores 0 — an extreme-breadth signal computed from the record, not missing data. The field distribution splits evenly (0.5/0.5) between machine learning/movement ecology and ML for ecology/computational sustainability, and topic counts spread across five clusters: advanced neural network applications (7), then wildlife ecology and conservation, video surveillance and tracking, domain adaptation and few-shot learning, and water-quality monitoring (4 each).

Collaboration71/100

Scores 71. The record lists 35 distinct coauthors — high for an early-career profile with 14 indexed works. The frequent-coauthor roster is anchored by movement-ecology and conservation researchers such as Francesca Cagnacci, Ellen O. Aikens, Meredith S. Palmer, and Jessica Kendall-Bar, alongside steady collaborators including Martin Becker, Nicholas J. Russo, and Will Rogers, matching the record's two-field split.

Activity67/100

Scores 67. Indexed output runs 2021: 1, 2022: 3, 2023: 1, 2024: 2, 2025: 2, 2026: 5 — 14 works total, with 9 in the last three years. The 2026 count of 5 is the record's highest and includes both Evolve-reviewed arXiv papers, indicating a recent, ongoing acceleration in posting activity.

Reproducibility60/100

Scores 60, averaged over 2 Evolve-reviewed papers. The two split: the active-learning biodiversity-survey paper rates high reproducibility readiness, while MoveBench rates low — Evolve notes that withheld data and code leave its headline findings unverifiable. With only two reviewed papers, the mean is indicative rather than stable.

Research summary

Justin Kay is a US-based researcher at MIT working at the intersection of machine learning and ecology. The OpenAlex-indexed record splits evenly (0.5/0.5) between machine learning/movement ecology and machine learning for ecology/computational sustainability, with topic clusters spanning advanced neural network applications (7 works) alongside wildlife ecology and conservation, video surveillance and tracking, domain adaptation and few-shot learning, and water-quality monitoring technologies (4 works each). The record counts 14 indexed works since 2021 — 1 (2021), 3 (2022), 1 (2023), 2 (2024), 2 (2025), 5 (2026) — with 9 works in the last three years, an h-index of 3 and an i10-index of 2: an early-career profile whose posting activity is accelerating. Two 2026 arXiv papers anchor the Evolve catalog: MoveBench, a first-author benchmark for global-scale probabilistic wildlife movement forecasting using energy scores and step-selection analysis (rated low reproducibility readiness because data and code were withheld), and an active-learning occupancy-modeling paper for AI-assisted biodiversity surveys (middle author, high reproducibility readiness; Evolve reports a 2–4× reduction in expert effort via Target-EIG review). Collaboration is broad — 35 distinct coauthors, including movement ecologists such as Francesca Cagnacci and Ellen O. Aikens and conservation scientist Meredith S. Palmer. This profile is name-resolved from public OpenAlex bibliographic data; no ORCID was provided.

Collaboration pattern

Collaboration is wide for an early-career record: 35 distinct coauthors (dimension score 71). The frequent-coauthor list is dominated by movement-ecology and conservation researchers — Francesca Cagnacci, Ellen O. Aikens, Meredith S. Palmer, Jessica Kendall-Bar — alongside steady collaborators such as Martin Becker, Nicholas J. Russo, and Will Rogers, consistent with the record's even split between ML and ecology fields.

Research keywords

wildlife movement forecastingtrajectory predictionprobabilistic forecastingbenchmark datasetsoccupancy modelingactive learningBayesian experimental designbiodiversity monitoringcamera trapsbioacousticsdomain adaptationfew-shot learning

Highlights

ACORN cuts expert review effort 2–4× in AI-assisted biodiversity surveys
On two biodiversity datasets spanning camera traps and bioacoustics, a score-calibrated occupancy model combined with Target-EIG active review reduces expert labeling effort by a reported 2–4×. Evolve rates the work solid while flagging that oracle labels are imperfect, and scores its reproducibility readiness high.
Bayesian experimental design decides which records experts review
Rather than sampling at random, the Target-EIG step applies Bayesian expected-information-gain selection to choose the continuous-score records whose labels are most worth acquiring, positioning active learning as a survey-design tool for occupancy estimation rather than a training-time trick.
MoveBench: a first benchmark for global-scale wildlife movement forecasting
This first-authored 2026 benchmark frames wildlife movement forecasting as probabilistic prediction, evaluated with energy scores and step-selection analysis. Evolve calls it a well-designed first benchmark but flags that withheld data and code leave headline findings unverifiable, rating reproducibility readiness low.

Timeline

2021First work appears in the OpenAlex-indexed record (1 publication).
2022Early peak with 3 indexed publications.
2024–2025Steady output of 2 indexed publications in each of 2024 and 2025.
2026Most active year: 5 indexed works, including both Evolve-cataloged papers — first-authored MoveBench and the active-learning biodiversity-survey paper.

Papers in the Evolve catalog

2026-09-22
Targeted Review for AI-Assisted Biodiversity Surveys: Active Continuous-Score Occupancy Modeling
2026-09-14first author
MoveBench: A Benchmark for Global-Scale Wildlife Movement Forecasting

Frequent co-authors

Shir Bar×1 Ellen O. Aikens×1 Martin Becker×1 Francesca Cagnacci×1 Juliet Cohen×1 Scott W. Forrest×1 Jessica Kendall-Bar×1 Madeleine Lucas×1 Macon Overcast×1 Meredith S. Palmer×1 Will Rogers×1 Nicholas J. Russo×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.

Bibliometric metrics come from an OpenAlex author snapshot only; no ORCID was provided, so name resolution relies on OpenAlex bibliographic matching and may merge or omit records for this name.
Citation indicators are early-career thin (h-index 3, i10-index 2); output and impact dimensions are therefore left unscored rather than estimated.
The reproducibility score (60) averages only 2 Evolve-reviewed papers — a small sample; the remaining indexed works have not been reviewed.
The 2026 publication count may include recently posted arXiv preprints not yet peer-reviewed.