Osvaldo Simeone

Northeastern University · United States

Unclaimed · AI-structured profileOpenAlex

Northeastern University researcher at the intersection of conformal prediction and wireless federated learning

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

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 data is insufficient to score: the OpenAlex record reports an h-index of 1 and i10-index of 1, plausibly a fragmented author profile, so total publication volume cannot be assessed. This axis would measure cumulative and per-year work counts.

Impact—

Impact cannot be scored: citation coverage is missing or unreliable, with the record showing only h-index 1 and i10-index 1, which do not credibly represent this researcher's influence. This axis would measure citations, citation trajectory, and uptake across venues.

Focus0/100

The focus score is 0, indicating a topically broad record rather than a concentrated one: OpenAlex topic clustering spreads across five areas — adversarial robustness in machine learning (12 works), reinforcement learning in robotics (5), explainable AI (4), advanced neural network applications (3), and advanced MIMO systems optimization (3) — while the field distribution splits evenly (0.5/0.5) between machine learning / signal processing and wireless communications / machine learning.

Collaboration39/100

The collaboration score is 39. The recent OpenAlex window lists 6 distinct coauthors, and Evolve's catalog identifies a stable repeat circle — Pietro Talli, Petar Popovski, Haifeng Wen, Nicolò Michelusi, Yang Yang, and Hong Xing. On the two reviewed 2026 papers Simeone appears in middle and last author positions without a corresponding-author role, consistent with contributing inside a compact, recurring team rather than a large distributed network.

Activity98/100

Activity is the strongest dimension at 98: counts-by-year attributes 28 works to 2026, and the three-year window totals 28 works (recent3yWorks 28), including both reviewed arXiv preprints. Earlier coverage is thin — the record shows a single 2009 entry — so the profile reads as a burst of intense current output atop sparse historical indexing.

Reproducibility55/100

The reproducibility score is 55, averaged over 2 reviewed papers (reproMean 55, reproPapers 2), both rated medium repro-readiness. Evolve's verdicts flag concrete gaps: the conformal-probing paper omits part of its proof and evaluates only a small synthetic MDP, while the over-the-air federated learning paper proves convergence only for uniform, not optimized, device selection.

Research summary

Osvaldo Simeone is a researcher at Northeastern University (US) whose recent record spans two equally weighted strands: machine learning / signal processing, centered on conformal prediction and reliable sequential decision-making, and wireless communications, centered on federated edge learning. Evolve's catalog includes two 2026 arXiv preprints. 'Risk-Aware Online Conformal State Probing' (last author) proposes a training-free action-and-probing scheme with a distribution-free missed-query guarantee; Evolve judges it promising but notes a partially omitted proof and evaluation limited to a small synthetic MDP. 'Non-Coherent Over-the-Air Federated Learning' (middle author) develops a non-coherent AirFL protocol using binary dithering, error feedback, and device scheduling with an O(1/√T) convergence bound, proven only for uniform device selection. Activity is the standout metric: 28 works in the recent three-year window (2026 alone accounts for 28), yielding an activity score of 98. Collaboration is modest (39), with 6 distinct coauthors and a recurring circle including Pietro Talli, Petar Popovski, Haifeng Wen, Nicolò Michelusi, Yang Yang, and Hong Xing. OpenAlex topic clustering is broad (adversarial robustness 12; reinforcement learning in robotics 5; XAI 4; neural network applications 3; MIMO optimization 3), giving a focus score of 0. Output and impact are unscored because the record's h-index (1) and i10-index (1) suggest a fragmented author profile; reproducibility averages 55 over the two reviewed papers.

Collaboration pattern

Collaboration is compact and recurring (score 39): the recent window lists 6 distinct coauthors, and Evolve identifies a stable circle of Pietro Talli, Petar Popovski, Haifeng Wen, Nicolò Michelusi, Yang Yang, and Hong Xing. On the two reviewed 2026 papers Simeone sits in middle and last author positions, neither as corresponding author, consistent with contributing within team-led projects across both the conformal-prediction and over-the-air federated learning lines.

Research keywords

online conformal predictionconformal risk controldistribution-free guaranteessequential decision-makingactive state probingover-the-air computationfederated learningnon-coherent detectionbinary ditheringerror feedbackdevice selectionwireless communications

Highlights

Distribution-free missed-query guarantees for online conformal probing
The 2026 preprint 'Risk-Aware Online Conformal State Probing' (last-author position) proposes a training-free action-and-probing scheme with a distribution-free missed-query guarantee for sequential decision-making. Evolve's verdict is promising but flags a partially omitted proof and evaluation restricted to a small synthetic MDP; reproducibility is rated medium.
Non-coherent over-the-air federated learning with O(1/√T) convergence
The 2026 preprint 'Non-Coherent Over-the-Air Federated Learning: Protocol, Convergence, and Device Scheduling' (middle-author position) designs a non-coherent AirFL protocol using binary dithering, error feedback, and device scheduling, with an O(1/√T) bound. Evolve notes convergence is proven only for uniform, not optimized, selection; reproducibility is rated medium.

Timeline

2009Earliest work recorded in the OpenAlex counts-by-year for this profile (1 work); coverage before 2026 is otherwise sparse.
202628 works are recorded for 2026 (recent3yWorks 28), the most active year in the record and the basis of the 98 activity score.
2026Released 'Risk-Aware Online Conformal State Probing' (arXiv), in the last-author position.
2026Released 'Non-Coherent Over-the-Air Federated Learning: Protocol, Convergence, and Device Scheduling' (arXiv), in a middle author position.

Papers in the Evolve catalog

Frequent co-authors

Pietro Talli×1 Petar Popovski×1 Haifeng Wen×1 Nicolò Michelusi×1 Yang Yang×1 Hong Xing×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 come from an OpenAlex author record (identity: openalex); the reported h-index of 1 and i10-index of 1 are inconsistent with a sustained career and likely reflect a fragmented or incomplete profile, so the output and impact axes are left unscored.
Coverage is sparse before 2026: counts-by-year shows only a single 2009 entry, so long-term productivity trends cannot be established.
Only two 2026 arXiv preprints from Evolve's catalog were reviewed; no peer-reviewed venue, funding, or award information is included.
OpenAlex topic labels (e.g., adversarial robustness, reinforcement learning in robotics) do not fully align with the two reviewed papers; the profile is name-resolved from public bibliographic data and author attribution may be noisy.