Urbashi Mitra

University of Southern California · United States

Unclaimed · AI-structured profileORCID OpenAlex

USC communication theorist, four-decade record (h-index 48), now driving identifiability theory for mixed causal DAGs

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-09-28) 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—

No score yet: the data needed for an output dimension was not aggregated in this snapshot. The axis measures publication volume and year-over-year productivity; per-year work counts exist here, but no composite output score was computed.

Impact—

No score: citation-impact data is insufficient in this snapshot. The axis measures citation volume and influence on later work; h-index (48) and i10-index (219) exist as raw figures, but no composite impact score was computed.

Focus100/100

Focus scores 100/100: the record's top five OpenAlex topics are all wireless-communication areas — Advanced Wireless Communication Techniques (132 works), Wireless Communication Networks Research (116), Cooperative Communication and Network Coding (94), Distributed Sensor Networks and Detection Algorithms (81), and Underwater Vehicles and Communication Systems (73). The field distribution also logs causal inference / statistical machine learning (1), matching the 2026 turn visible in the reviewed papers.

Collaboration28/100

Scoring 28/100, collaboration is the lowest recorded axis: the snapshot lists only 3 distinct coauthors, with Sambit Mishra, Yingying Wang, and Christine K. Johnson recurring as frequent collaborators. On both reviewed 2026 papers Mitra is the last author without a corresponding-author flag, a pattern consistent with senior supervisory authorship. Given a record running from 1989 to 2026, the low distinct-coauthor count more plausibly reflects coverage limits in OpenAlex disambiguation than the true network size.

Activity100/100

Activity scores 100/100 on an exceptionally durable record: works are logged in every year from 1989 to 2026, recent annual counts run 27 (2024), 16 (2025), and 19 (2026), and the most recent three-year window holds 62 works. The single-year peak is 39 works in 2014, inside a 2010–2018 stretch that never falls below 16.

Reproducibility100/100

Reproducibility scores 100/100, with the mean computed from just one reviewed paper (reproPapers = 1). That paper — on identifiability of mixed ordinal–exponential causal DAGs under linear parametric models — carries Evolve's high reproducibility-readiness rating, while its 2026 companion is rated medium. The perfect score is best read as promising early evidence rather than a stable, multi-paper estimate.

Research summary

Urbashi Mitra is based at the University of Southern California, with a bibliometric record resolved via OpenAlex against ORCID 0000-0002-8896-1177 that spans nearly four decades. The snapshot reports an h-index of 48 and an i10-index of 219, works logged in every year from 1989 through 2026, and 62 works in the most recent three-year window. OpenAlex topic modeling shows a record dominated by wireless communication: Advanced Wireless Communication Techniques (132 works), Wireless Communication Networks Research (116), Cooperative Communication and Network Coding (94), Distributed Sensor Networks and Detection Algorithms (81), and Underwater Vehicles and Communication Systems (73). The most recent output marks a visible turn toward causal inference and statistical machine learning: Evolve reviewed two 2026 arXiv preprints on the identifiability of mixed ordinal–exponential-family causal DAGs, on both of which Mitra is the last author. Evolve's verdicts credit the pair with rigorous identifiability theory — every-parameter results with sharp converses, and generic direction identifiability for ordinal–exponential edges — while flagging synthetic-only validation and, on one paper, missing baselines and real data. Radar dimensions score activity, focus, and reproducibility at 100/100 each, with collaboration at 28 based on three distinct coauthors recorded in this snapshot (frequent collaborators: Sambit Mishra, Yingying Wang, Christine K. Johnson). Composite output and impact scores were not computed for this snapshot.

Collaboration pattern

In this snapshot the collaboration dimension scores 28/100 with only three distinct coauthors recorded; Sambit Mishra, Yingying Wang, and Christine K. Johnson recur as frequent collaborators. On both reviewed 2026 papers Mitra is the last author without a corresponding-author flag, consistent with a senior supervisory role. Given the length of the record, the low distinct-coauthor count almost certainly reflects coverage limits in OpenAlex disambiguation rather than the true size of the collaboration network.

Research keywords

Wireless communication systemsWireless networksCooperative communicationNetwork codingDistributed sensor networksDetection algorithmsUnderwater communicationCausal discoveryDAG identifiabilityOrdinal dataExponential familyMarkov equivalence class

Highlights

Every-parameter identifiability for mixed ordinal–exponential causal DAGs
A 2026 arXiv preprint with Mitra as last author proves every-parameter identifiability of mixed ordinal–exponential causal edges under linear parametric models, with sharp converses. Evolve rates the theory rigorous but notes validation is synthetic-only; reproducibility readiness is high.
Generic identifiability of edge directions in epidemiological causal graphs
A 2026 arXiv preprint with Mitra as last author proves generic identifiability of ordinal–exponential edge directions in mixed DAGs, targeting epidemiological data. Evolve notes the theory is bivariate-only and experiments lack baselines and real data; reproducibility readiness is medium.
Four decades of unbroken publication output
The OpenAlex record logs works in every year from 1989 through 2026, peaking at 39 works in 2014 and totalling 62 works over the most recent three-year window, with 27 (2024), 16 (2025), and 19 (2026) in the latest counts.

Timeline

1989Earliest work in the snapshot (1 publication); the record continues unbroken every year through 2026.
2004Output reaches 32 works, part of a sustained early-2000s stretch with annual counts running from the high teens to the low thirties.
2014Single-year peak of the record: 39 works, within a 2010–2018 run whose annual counts stay between 16 and 39.
2024Late-career output stays strong: 27 works in 2024 and 16 in 2025, contributing to 62 works across the most recent three-year window.
2026Two reviewed arXiv preprints on causal-DAG identifiability appear with Mitra as last author — a marked turn toward causal inference; 19 works are already logged for the year.

Papers in the Evolve catalog

Frequent co-authors

Sambit Mishra×2 Yingying Wang×1 Christine K. Johnson×1

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 derive from an OpenAlex snapshot only, with identity resolved against ORCID 0000-0002-8896-1177; name disambiguation may merge or split records.
Composite output and impact dimension scores were not computed for this profile, even though raw citation aggregates (h-index 48, i10-index 219) are present.
The reproducibility score of 100 rests on a single reviewed paper (reproPapers = 1) and is early-stage evidence, not a stable estimate.
Item-level Evolve review covers only two 2026 arXiv preprints in causal inference; the long wireless-communication record is reflected solely through OpenAlex topic counts.