KAIST machine learning researcher bridging generative modeling on manifolds and ML-driven experimental design in biology
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
12
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—
Output scoring is unavailable: the profile lacks a verified total publication count needed for this axis. The yearly counts on file (1 work in 2018 rising to 13 in 2026) cover only the OpenAlex-indexed subset, so no score is assigned.
Impact—
Impact scoring is not yet computed for this profile. Partial signals are on file — h-index 12 and i10-index 12 — but the citation-based, field-normalized measures this axis requires are incomplete, so no score is assigned yet.
Focus0/100
Focus scores 0/100, reflecting breadth rather than concentration: the field distribution splits evenly (0.5 computational biology / ML for experimental design, 0.5 machine learning), and topic clustering spans at least five areas, led by multiple sclerosis research (21) and machine learning in healthcare (14).
Collaboration51/100
Collaboration scores 51/100, with 12 distinct co-authors on file. Recurring partners span both of this profile's fields — including Gabriele Scalia, Carl Edwards, Xiner Li, Sara Mostafavi, and Ehsan Hajiramezanali — indicating steady but not large-team co-authorship; on the two reviewed 2026 papers this profile holds middle-author, non-corresponding positions.
Activity100/100
Activity scores 100/100, the highest axis on file. Indexed works rise from 1 in 2018 to 8 in 2020, hold at 10 in 2021–2022, stay near 9 per year through 2025, and reach 13 in 2026, with 31 works in the last three years — continuous output across nine consecutive years.
Reproducibility55/100
Reproducibility scores 55/100, a mean taken over the 2 papers Evolve has reviewed; both are rated medium repro-readiness. Methods are benchmark-backed (e.g., a 1,389-screen CRISPR benchmark), but the moderate score suggests reproduction depends on filtered test sets and nontrivial setup.
Research summary
Edward De Brouwer is a machine learning researcher at the Korea Advanced Institute of Science and Technology (KAIST) whose indexed portfolio splits evenly between machine learning and computational biology, specifically machine learning for experimental design. The OpenAlex-resolved record (ORCID 0000-0003-0608-0155) shows steady growth from 1 indexed work in 2018 to a plateau of 8–10 works per year through 2025 and 13 in 2026, with 31 works in the most recent three-year window and an h-index of 12 (i10-index 12). Topic clustering is led by multiple sclerosis research studies (21), machine learning in healthcare (14), topological and geometric data analysis (10), COVID-19 clinical research (9), and computational drug discovery (9) — a health-oriented machine learning portfolio. Two 2026 arXiv papers reviewed by Evolve mark the current research front: one develops flow matching on the Wasserstein space of Riemannian manifolds with supporting theory and strong benchmarks; the other introduces a 1,389-screen CRISPR benchmark and AssayLoop, reaching 5.67× random enrichment factor for adaptive hit discovery. On both, this profile is recorded as a middle author, non-corresponding.
Collaboration pattern
Twelve distinct co-authors are on file, matching the 12-name frequent-coauthor list; the collaboration score is 51/100. The network spans both recorded fields — machine learning and computational biology — with recurring partners including Gabriele Scalia, Carl Edwards, Xiner Li, Ehsan Hajiramezanali, Namkyeong Lee, Sara Mostafavi, and Rex Ying. On the two Evolve-reviewed 2026 papers this profile appears as a middle author and not corresponding, consistent with a stable mid-size co-author network rather than solo work or large-consortium writing.
Research keywords
Highlights
Flow matching moves onto Wasserstein space
A 1,389-screen benchmark for adaptive CRISPR hit discovery
One portfolio, two fronts: transport theory meets experimental design
The two reviewed 2026 papers sit at opposite ends of one program — geometric generative modeling via optimal transport on Riemannian manifolds, and ML-driven adaptive experimental design for CRISPR screens — matching the recorded 50/50 field split between machine learning and computational biology.
Timeline
- 2018First indexed work appears in the OpenAlex record (1 work).
- 2020Output ramps up to 8 indexed works, beginning a sustained multi-year run.
- 2021Indexed output reaches 10 works, held again in 2022 — the first plateau.
- 2023Steady 9-work years run through 2023–2025, alongside 12 distinct co-authors on file.
- 2026Most prolific year on record with 13 indexed works; two Evolve-reviewed arXiv papers appear — Wasserstein-space flow matching and the 1,389-screen CRISPR benchmark with AssayLoop.
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.
- Bibliometrics come from an OpenAlex identity record only (ORCID 0000-0003-0608-0155); name disambiguation relies on that public profile.
- Output and impact axis scores are missing; only partial signals (h-index 12, i10-index 12) are available, with no per-paper citation counts on file.
- The reproducibility mean (55) is computed from only 2 Evolve-reviewed papers and may not reflect the wider portfolio.
- Topic labels and yearly counts describe the indexed corpus; non-indexed outputs such as code, datasets, and theses are not captured.