Signal processing researcher working on adaptive filtering, sparse sensing, and Gaussian-process machine learning.
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
24
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—
Not yet scored: complete publication-volume records are unavailable for this profile. This axis would measure total scholarly output and its growth over time, complementing the 51 works already counted in the last three years.
Impact—
Not yet scored: citation-linked records covering this researcher's full body of work are incomplete. This axis would measure citation-based influence, for example uptake of adaptive-filtering and sparse-sensing methods, once comprehensive citation data is available.
Focus0/100
The focus score is 0, reflecting an even split (0.5/0.5) between the two indexed fields — machine learning/statistical signal processing centered on Gaussian processes, and image processing/compressed sensing. Topic counts reinforce this breadth: advanced adaptive filtering (96), sparse and compressive sensing (63), neural networks (42), control systems and identification (29), and optimization and variational analysis (26). This is a deliberately wide portfolio rather than a narrow single-topic agenda.
Collaboration32/100
The collaboration score is 32, based on 4 distinct co-authors in the snapshot, with Emilio Ruiz-Moreno, Baltasar Beferull-Lozano, and Shiwen An recurring most often. Both reviewed 2026 papers list him in middle or last authorship without corresponding-author flags, consistent with a senior contributor embedded in a small, stable circle rather than a large distributed network.
Activity100/100
The activity score is the maximum 100. Publication counts in the snapshot rise from single digits through the 2010s to peaks of 20 works in 2023 and 2024, 19 in 2025, and 12 already in 2026, with 51 works in the most recent three-year window. This sustained recent volume places him in the most active tier, spanning arXiv preprints and indexed venues.
Reproducibility78/100
The reproducibility score is 78, averaged over the 2 Evolve-reviewed papers. The warped Gaussian-process gradient recursion is rated high for reproducibility readiness thanks to its exact recursion, though validation rests on one small synthetic task; the quantum-inspired tensor-network inpainting paper is rated medium, with single-seed evaluation and an omitted proof tempering its claims.
Research summary
Konstantinos Slavakis is a researcher at Institute of Science Tokyo whose profile is name-resolved from public OpenAlex bibliographic data and sits at the intersection of statistical signal processing and machine learning. His indexed output reaches back to 2001 and shows a pronounced late-career acceleration: yearly counts climb from single digits through the 2010s to peaks of 20 publications in 2023 and 2024, 19 in 2025, and 12 already in 2026, with 51 works in the most recent three-year window. Topic counts concentrate on advanced adaptive filtering techniques (96), sparse and compressive sensing (63), neural networks and applications (42), control systems and identification (29), and optimization and variational analysis (26), splitting evenly between two indexed fields — machine learning/statistical signal processing centered on Gaussian processes, and image processing/compressed sensing. The snapshot reports an h-index of 24 and an i10-index of 38, while the output and impact radar axes remain unscored. Two 2026 arXiv papers reviewed by Evolve — quantum-inspired trainable tensor networks for image inpainting and an exact online gradient recursion for warped Gaussian-process transformations — cast him as a senior author bridging classical estimation theory with kernel-based and quantum-inspired learning; the latter is rated high for reproducibility readiness (mean 78 across 2 reviewed papers).
Collaboration pattern
The snapshot lists only 4 distinct co-authors, giving a collaboration score of 32 and pointing to a compact, durable working circle. Emilio Ruiz-Moreno, Baltasar Beferull-Lozano, and Shiwen An recur most frequently, and both 2026 reviewed papers place him as a non-corresponding middle or last author — a senior-contributor pattern in which he supplies theoretical machinery while collaborators drive experiments.
Research keywords
Highlights
Exact online gradient recursion for warped Gaussian processes
Quantum-inspired tensor networks for parameter-efficient image inpainting
Late-career publication surge to record output
Snapshot counts rise to a record 20 publications in 2023 and again in 2024, 19 in 2025, and 12 already in 2026, with 51 works over the most recent three years — underpinning the maximum activity score of 100.
Timeline
- 2001Earliest indexed output in the snapshot: 2 works recorded in 2001.
- 2010Output reaches an early peak of 14 indexed works in 2010.
- 2023Publication counts hit the record high of 20 works in 2023, a level repeated in 2024.
- 2026Two Evolve-reviewed arXiv papers appear: quantum-inspired tensor networks for image inpainting (last author) and online gradient computation for warped Gaussian-process transformations (middle author).
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 are name-resolved from OpenAlex only; the frequent-coauthor list includes the researcher's own name, indicating residual entity-resolution noise.
- The output and impact radar axes are unscored because records are incomplete, even though the snapshot reports h-index 24 and i10-index 38.
- The reproducibility score of 78 averages only the 2 Evolve-reviewed arXiv preprints, so it is an early-stage signal rather than a stable measure.
- Yearly counts include partial 2026 data (12 works as of the snapshot), so the latest year is not yet complete.