CV

Boston, MA · Open to relocation

Statistical Machine Learning · Model Evaluation
EEG & BCI · Research Software

Ph.D. in Computational Sciences with research in evaluating and adapting models when data changes. My work combines chronological evaluation, confidence-calibrated source screening, continual EEG adaptation, and reproducible Python/R software. I measure predictive gains alongside computational cost and individual performance loss. I have experience in independent university mathematics teaching, assessment design, and undergraduate research mentoring. EEG is my primary research domain; my prior industry work involved sensor data, MySQL, monitoring dashboards, and automated reporting.

Extended Research & Software CV · 4 pages Short Resume · 2 pages Email Yiming LinkedIn

Research and teaching on this page updated 6 Oct 2026. Downloadable CV and short resume dated 20 Sep 2026. View the complete current research list · View teaching experience.

Methods and Software

  • Time-series classification, model evaluation, and distribution-shift analysis
  • CSP/FBCSP, covariance features, tangent-space methods, and Euclidean alignment
  • Domain adaptation, transfer learning, chronological and nested validation, and shared-target bootstrap uncertainty
  • Continual EEG adaptation, history replay, and participant-level retention analysis
  • Python, R, SQL, NumPy, SciPy, pandas, MNE-Python, statsmodels, Git, and Linux

Selected Evidence

  • First-author Frontiers paper: published 28 Aug 2026, volume 20, Article 1895016
  • Research software: CrossDA, ShiftDx, and NeuroStream
  • EEG competition: Kaggle HMS Silver Medal, rank 98 of 2,767 teams (top 4%)
  • Industry: two years of IoT data engineering before doctoral research
  • Teaching: instructor of record for precalculus and calculus at UMass Boston, 2022–2024; undergraduate research co-mentoring, 2025–2026

Selected Research

  1. A Confidence-Gated Source Selection Strategy for Cross-Session Transfer in Brain-Computer Interfaces
    Published · Frontiers in Human Neuroscience - Brain-Computer Interfaces · Yiming Shen and David Degras. doi:10.3389/fnhum.2026.1895016
  2. Confidence-Calibrated Source Screening for Cross-Session Motor-Imagery EEG under Distribution Shift
    Working Paper · Yiming Shen.
  3. LOSO versus Chronological Validation in Longitudinal BCI
    Under Review · IEEE Access · Yiming Shen.
  4. Cross-Channel Prediction, Classification, and Individual Retention in Low-Channel EEG Adaptation
    Working Paper.
  5. User Linkage and Replay Utility in BCI
    Working Paper · Yiming Shen and Xinmin Chu.
  6. Small Calibration Cohorts Do Not Reliably Select Motor-Imagery EEG Systems for New Participants
    Under Review · Scientific Reports · Yiming Shen.

Teaching Experience

Instructor of Record

University of Massachusetts Boston
2022–2024 (multiple semesters)

  • MATH 130: Precalculus · Undergraduate
  • MATH 140: Calculus I · Undergraduate
  • Prepared and delivered lectures, assigned homework, and supported exam preparation.
  • Designed assessments and graded student work.

Teaching Assistant

University of Massachusetts Boston
Fall 2020–Fall 2022

  • MATH 130: Precalculus · Undergraduate
  • MATH 140: Calculus I · Undergraduate
  • MATH 345: Probability and Statistics · Undergraduate

Teaching Assistant

Valparaiso University
Spring 2017–Spring 2018

  • STAT 340: Statistics for Decision Making I · Undergraduate
  • STAT 540: Statistics for Decision Making II · Graduate

Student Research Mentoring

Graduate Student Mentor — Undergraduate Research

STEMPOWER Research Fellowship in Mathematics, University of Massachusetts Boston
2025–2026 academic year

I co-mentored a third-year undergraduate alongside faculty mentor David Degras on classifying left- and right-hand motor imagery from EEG.

  • Met twice weekly throughout the academic year.
  • Taught classification theory and set up the computational environment for code testing and benchmark replication.

The student reproduced a complete benchmark and prepared a poster displayed on campus.

Selected Research Software

CrossDA

CLI and Python API for comparing pooling, weighting, and source-selection strategies across EEG sessions, with reproducible experiments and auditable outputs.

ShiftDx

Offline fixed-reference MI-EEG drift analysis across five shift metrics and no-adaptation, domain-adaptation, and retraining conditions.

NeuroStream

Sample-paced pseudo-online replay of motor-imagery EEG with CSP, FBCSP, and tangent-space pipelines, Euclidean alignment, and progressive inference visualization.