RL researcher at the Finnish Defence Forces working on safe exploration and novelty-driven reinforcement learning
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
0
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 snapshot records an h-index of 0 and no i10-index, with all indexed works dated 2026. This axis would measure total publication volume, authorship positions, and venue quality once a fuller record accumulates.
Impact—
Not yet scored: citations, h-index and i10-index are all zero in the OpenAlex snapshot, reflecting works published only in 2026. This axis would measure citation influence and scholarly uptake once the record matures.
Focus0/100
Scores 0 on topical concentration: the OpenAlex record spreads across five topics, led by Reinforcement Learning in Robotics (3 works) with one work each in distributed control of multi-agent systems, advanced control optimization, fault detection, and AI-based planning. The field split is 50% machine learning (reinforcement learning) and 50% safe reinforcement learning, though the two Evolve-reviewed papers are closely related exploration/safety studies.
Collaboration39/100
Scores 39, a modest collaboration footprint: 6 distinct coauthors appear in the record, and the frequent-coauthor list names Wenyan Yang, Arsenii Mustafin, Joni Pajarinen, Simone Parisi, Shreeram Murali, and Shankar A. Deka. On both Evolve-reviewed 2026 papers Baumann is a contributing co-author — last author on the safe-exploration paper and middle author on SUN — rather than first or corresponding author.
Activity52/100
Scores 52, mid-range: the snapshot counts 5 works in the last three years, with countsByYear showing all 5 appearing in 2026 and nothing earlier, consistent with an emerging record. Both catalog papers are 2026 arXiv preprints, so recent output is real but recent and thin; sustained multi-year production is not yet visible in the data.
Reproducibility55/100
Scores 55, based on Evolve's reproducibility assessment of 2 papers (mean 55). Both are rated medium readiness. The safe-exploration paper is flagged for results resting on single runs and a retuned baseline despite high-probability safety proofs; SUN earns credit for strong benchmarks and honest ablations, though its proofs are oracle-scoped and baselines few.
Research summary
Dominik Baumann is a Finland-based researcher affiliated with the Finnish Defence Forces whose indexed record, resolved from OpenAlex (no ORCID linked), centers on reinforcement learning. The snapshot shows an emerging profile: h-index 0 and i10-index 0, five works in the last three years, and countsByYear listing all five in 2026 with nothing earlier. The record splits evenly — 50% machine learning (reinforcement learning) and 50% safe reinforcement learning — and spans five OpenAlex topics, led by Reinforcement Learning in Robotics (3 works) with single works in distributed control of multi-agent systems, advanced control optimization, fault detection, and AI-based planning. Evolve has reviewed two 2026 arXiv papers. In 'Computationally efficient safe exploration in reinforcement learning' (last author) the authors claim O(1) kernel-estimator safe RL for constrained MDPs with high-probability safety proofs, which Evolve judges promising but dependent on single runs and a retuned baseline. In 'SUN: Reaching for Novelty in Reinforcement Learning' (middle author) the team proposes a novelty-by-reachability goal score for exploration, praised for strong benchmarks and honest ablations though its proofs are oracle-scoped and baselines few. Both papers carry medium reproducibility readiness, averaging 55. Collaboration is modest (score 39, six distinct coauthors), with frequent coauthors including Wenyan Yang, Joni Pajarinen, and Simone Parisi. This is an early-stage profile; output and impact dimensions lack sufficient data.
Collaboration pattern
The record shows 6 distinct coauthors and a collaboration score of 39, indicating modest, small-team work rather than large consortium output. Frequent coauthors are Wenyan Yang, Arsenii Mustafin, Joni Pajarinen, Simone Parisi, Shreeram Murali, and Shankar A. Deka. On both reviewed 2026 papers Baumann contributes as a co-author — last author on the safe-exploration study and middle author on SUN — not as first or corresponding author.
Research keywords
Highlights
O(1) kernel-estimator safe exploration for constrained MDPs
SUN: novelty × reachability goal scoring for RL exploration
Timeline
- 2026OpenAlex countsByYear records 5 works in 2026 — the only year with activity in the snapshot — placing Baumann in the recent-3y window with an h-index of 0.
- 2026arXiv preprint 'Computationally efficient safe exploration in reinforcement learning' appears, with Baumann in the last author position.
- 2026arXiv preprint 'SUN: Reaching for Novelty in Reinforcement Learning' appears with Baumann as a middle author; Evolve reviews both 2026 papers, rating each medium for reproducibility readiness.
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.
- Identity is name-resolved from public OpenAlex bibliographic data only; no ORCID is linked, so name collisions with other researchers are possible.
- Bibliometric scores come from an OpenAlex snapshot; the citation record (h-index 0, i10-index 0) covers only works dated 2026.
- Output and impact dimensions are unscored because the underlying data is insufficient in the snapshot.
- Only two papers, both 2026 arXiv preprints, have been reviewed by Evolve; no peer-reviewed venue record appears in the data.