University of Florida researcher focused on neural, error-bounded compression of scientific data
2
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
1
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-17) 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 record behind this profile indexes too few works to compute reliable output metrics such as publication volume and i10-index (currently 0). This axis would measure total and per-year research output once disambiguation improves.
Impact—
Not yet scored: citation-based impact cannot be assessed — the record lists an h-index of 1 and i10-index of 0 with sparse citation data. This axis would measure citation counts, citation-derived indices, and influence on the field.
Focus100/100
Scores at the maximum (100/100): indexed topics concentrate on a tight cluster — traffic control and management (5 works), video coding and compression technologies (4), advanced data compression techniques (4), traffic and road safety (3), and algorithms and data compression (3) — with the reviewed 2026 papers both squarely in scientific data compression, the single field in fieldDist.
Collaboration28/100
Scores low (28/100): the OpenAlex record lists only 3 distinct coauthors alongside 11 works in the recent three-year window, indicating a small, stable collaboration circle rather than a broad network. Recurring partners in the Evolve catalog are Liangji Zhu, Anand Rangarajan, and Surya Majumder, consistent with compact multi-author arXiv papers rather than large consortium-style teams.
Activity72/100
Scores solidly (72/100): the record shows 11 works in the recent three-year window, and countsByYear attributes all 11 to 2026, signaling an active current phase. The two Evolve-reviewed arXiv papers both date from 2026. Earlier indexed activity is sparse (1 work in 1993, 2 in 1997, 1 in 2012), so the activity score reflects this concentrated recent burst.
Reproducibility55/100
Scores mid-range (55/100), based on 1 reproducibility-assessed paper with a mean readiness of 55. Evolve rates the 2026 neural residual compression paper as medium readiness: results are reported mainly in figures, classical baselines are absent, and error-bound guarantees are described as thin — reproducible in outline, but with gaps a reimplementation would need to close.
Research summary
Sanjay Ranka is a researcher at the University of Florida whose recent work centers on scientific data compression with machine learning. This profile is name-resolved from public bibliographic data (OpenAlex): the underlying record lists an h-index of 1 and an i10-index of 0 across a small set of indexed works spanning 1993 to 2026, so output and impact cannot be scored, while observed activity is heavily concentrated in 2026, to which countsByYear attributes all 11 recent works. Focus is exceptionally tight (100/100): indexed topics cluster around compression and traffic themes — traffic control and management (5 works), video coding and compression technologies (4), advanced data compression techniques (4), traffic and road safety (3), and algorithms and data compression (3) — and the single field in fieldDist is scientific data compression via machine learning. The Evolve catalog holds two 2026 arXiv papers. The first, with middle authorship, proposes a residual-vector-quantization, pixel-U-Net, and Guaranteed Autoencoder pipeline reported to cut NRMSE at equal compression ratio on three datasets; Evolve flags figure-only results, missing classical baselines, and a thin guarantee, rating reproducibility medium (55/100 from one assessed paper). The second, in the last authorship position, extends learned compression to multivariate data through cross-variable latent decorrelation and autoregressive entropy modeling. Frequent coauthors are Liangji Zhu, Anand Rangarajan, and Surya Majumder, within a small three-person collaboration circle.
Collaboration pattern
The record shows a compact collaboration pattern: 3 distinct coauthors across 11 works in the recent three-year window, matching a low collaboration score (28/100). The Evolve catalog repeatedly features Liangji Zhu, Anand Rangarajan, and Surya Majumder, suggesting a small, stable working group on learned scientific data compression. Authorship positions in the two reviewed papers — middle and last — point to a contributor role on joint arXiv work rather than a large-team or consortium style.
Research keywords
Highlights
First RVQ + pixel-U-Net + Guaranteed Autoencoder pipeline for error-bounded scientific compression
Learned compression for multivariate scientific data
A second 2026 arXiv paper, with Ranka in the last author position, addresses multivariate scientific data compression through learned cross-variable latent decorrelation and autoregressive entropy modeling. It appears in the Evolve catalog without a verdict, keywords, or reproducibility rating yet, so its claims remain to be assessed.
A concentrated 2026 research push on compression
countsByYear attributes all 11 recent works to 2026, and the focus dimension scores 100/100 with topics clustering on data compression, video coding, and traffic-related themes — evidence of a tightly scoped, currently very active, compression-centric research program.
Timeline
- 1993Earliest indexed work in the OpenAlex record (1 work indexed).
- 19972 works indexed in the record this year, the highest early-year count.
- 20121 work indexed — an isolated year between the sparse early records and the current phase.
- 2026Most active year: 11 works indexed, including both Evolve-reviewed arXiv papers on scientific data compression.
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 OpenAlex only; an h-index of 1 and i10-index of 0 over a record spanning 1993–2026 suggest possible incomplete author disambiguation, so output and impact remain unscored.
- Only 2 papers from this researcher are in the Evolve catalog, and reproducibility was assessed for just 1 (reproPapers = 1); the reproducibility score rests on a single paper.
- No ORCID was provided; identity resolution relies solely on the OpenAlex record.
- Topic labels and counts reflect OpenAlex automated topic assignment and may not match the researcher's own framing of their fields.