ML researcher at Eindhoven University of Technology: classification, few-shot learning, and uncertainty quantification
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
38
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-16) 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 underlying output data is insufficient in this snapshot. This axis would measure cumulative publication volume, per-year productivity, and venue distribution, none of which can be fully verified from the present bibliometric feed.
Impact—
Not yet scored: although a headline h-index exists in the raw feed, the citation-graph data needed to score influence is incomplete in this snapshot. This axis would measure citation volume, citation density, and uptake by follow-up work.
Focus0/100
Scores 0, reflecting a broad rather than concentrated footprint: the field distribution splits exactly 50/50 between machine learning and machine learning (uncertainty quantification), and the top five topic clusters — Machine Learning and Data Classification (123 works), Domain Adaptation and Few-Shot Learning (50), Scientific Computing and Data Management (43), Machine Learning and Algorithms (40), and Advanced Neural Network Applications (33) — span five distinct directions instead of one narrow line of work.
Collaboration46/100
Scores 46, a mid-range value grounded in 9 distinct recent coauthors. Frequent collaborators named in the snapshot include Abhash Kumar Jha, Diana Alexandra Onuţu, Neeratyoy Mallik, Swagatam Haldar, Sam Laing, Niccolò Ajroldi, Shiwei Liu, Aaron Klein, and Fangqin Zhou; both reviewed 2026 papers are team efforts with this researcher in non-corresponding roles (last author and middle author respectively).
Activity100/100
Scores 100, the maximum: 56 works in the most recent three-year window and an annual series that rose from 15 works in 2021 to a 2022 peak of 34, then stayed high at 14 (2023), 19 (2024), 15 (2025), and 22 already in 2026 — a sustained cadence with no recorded gap since the series begins in 1996.
Reproducibility55/100
Scores 55, the mean over the 2 papers with reproducibility assessments, both rated medium readiness. Evidence depth differs between them: the coregionalisation diagnostic claims 16-benchmark support with delimited scope, while the scaling-law amortisation paper is backed only by replay evidence and strong results on 4 of 5 grids, which tempers its verification profile.
Research summary
Joaquin Vanschoren is a machine learning researcher affiliated with Eindhoven University of Technology whose bibliometric record, as resolved from OpenAlex, shows a sustained and steeply rising publication profile. The snapshot credits an h-index of 38 with 75 papers carrying at least ten citations (i10 = 75), and 56 works in the most recent three-year window; yearly output climbed from 15 works in 2021 to a 2022 peak of 34, then held at 14-22 works through 2026, giving the activity dimension a maximum score of 100. Topic modelling concentrates on Machine Learning and Data Classification (123 works), Domain Adaptation and Few-Shot Learning (50), Scientific Computing and Data Management (43), Machine Learning and Algorithms (40), and Advanced Neural Network Applications (33). The two papers Evolve has reviewed are both 2026 arXiv preprints: a diagnostic study of when Gaussian-process coregionalisation improves joint negative log-likelihood, where residual correlation is claimed to predict gains across 16 benchmarks, and a Bayesian-optimisation approach to amortising scaling-law data collection with claimed 10-100x compute savings that currently rests on replay evidence alone. Both are judged medium in reproducibility readiness, matching a reproducibility score of 55 over 2 assessed papers. Collaboration scores mid-range (46) on 9 distinct recent coauthors.
Collaboration pattern
The collaboration profile scores 46 on a base of 9 distinct recent coauthors. Frequent collaborators named in the snapshot include Abhash Kumar Jha, Diana Alexandra Onuţu, Neeratyoy Mallik, Swagatam Haldar, Sam Laing, Niccolò Ajroldi, Shiwei Liu, Aaron Klein, and Fangqin Zhou. Both reviewed 2026 preprints are multi-author efforts in which this researcher holds a non-corresponding position (last author on the GP coregionalisation study, middle author on the scaling-law amortisation paper), a senior-contributor rather than lead-correspondent pattern.
Research keywords
Highlights
Residual correlation as a predictor of GP coregionalisation gains
Amortising the cost of building scaling laws
Timeline
- 1996Earliest year in the OpenAlex publication series, with 1 recorded work.
- 2015Output reached 17 recorded works, an early-series high before later fluctuations.
- 2021Output reached 15 recorded works, opening the current high-productivity phase.
- 2022Peak productivity year with 34 recorded works in the OpenAlex series.
- 202622 works recorded to date, including both Evolve-reviewed arXiv preprints (GP coregionalisation diagnostics; scaling-law cost amortisation).
Papers in the Evolve catalog
Frequent co-authors
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 scores (activity, collaboration, focus, reproducibility) are computed from OpenAlex data only; identity is resolved via OpenAlex.
- Output and impact dimensions are unscored because the underlying data is insufficient in this snapshot.
- Only 2 papers from the Evolve catalog were reviewed; highlights and the reproducibility score reflect that small sample.
- No country or affiliation-history data was provided beyond the current institution string.