Andrew Ma

Student Researcher & Machine Learning Intern

Tesla STEM High School

University of Washington (UW)

About Me

I am a student researcher at Tesla STEM High School and a Machine Learning Research Intern at the University of Washington (UW). My research centers on developing principled, resource-efficient deep learning architectures for scientific discovery and high-impact real-world systems.

In addition to my research, I serve as a Web Developer for the Competitive Programming Initiative (CPI), an Ambassador for the USAIIAO (USA International Artificial Intelligence Olympiad), and a certified Sno-King Ice Hockey Referee.

I also write about machine learning, model efficiency, and engineering systems on daily.dev.

Research Interests

I focus on designing and deploying principled machine learning models—spanning Vision Transformers (ViTs), Graph Neural Networks (GNNs), long-sequence temporal attention, efficient token sparsification, and dense context compression to address pressing challenges in climate resilience & precision agriculture, watershed & ecological forecasting, remote sensing & satellite observation, and clinical healthcare monitoring.

news

Nov 01, 2026	Our two physics research papers were accepted to the Spring 2026 APS Northwest Section Meeting (APS PNW) (presented May 21, 2026): PSTL: Investigating Particle Scattering in Triply Periodic Minimal Surface Lattices (Session N01.07) and Quantifying Non-Gaussian Scattering Tails in Micro-Heterogeneous Phantoms for Proton Therapy Applications (Session P01.01).
Oct 31, 2026	Excited to announce that our expanded paper “CalAD: A Calibrated MRI Baseline for Alzheimer’s Detection on OASIS-3” (co-authored with Khush Mehta) has been accepted as a poster to SPIE Medical Imaging 2027!
Oct 27, 2026	Our paper “Beyond Site Decodability: When Does a Scanner Shortcut Actually Inflate Diagnostic Performance?” (solo author) concluded review for the ML4H 2026 Findings track. While the paper was not accepted in the final decision, it received encouraging reviews—scores of 4 (Marginal Accept), 3 (Marginal Reject), 4 (Marginal Accept), 4 (Marginal Accept), and 3 (Marginal Reject)—with reviewers highlighting the importance of the Shortcut Gap metric. Our extensive rebuttal and new analyses across corrected IXI ($N=563$) and ABIDE I cohorts successfully addressed key methodological feedback, demonstrating that the gap serves as a sensitive screening diagnostic that scales from 0.074 to 0.240 under induced site-label correlation. We thank the reviewers and program chairs on OpenReview as we prepare the extended manuscript for future submission!

selected publications

ACM GeoAI

CONFLUENCE: Does Cross-Scale Graph Attention Help Watershed Vegetation Forecasting?

Andrew Ma

In 9th ACM SIGSPATIAL International Workshop on AI for Geographic Information Systems (ACM GeoAI 2026), 2026

Accepted to ACM SIGSPATIAL GeoAI 2026. Also submitted to NeurIPS TCCML Workshop. Solo author.

Bib HTML

@inproceedings{ma2026geoai,
  title = {CONFLUENCE: Does Cross-Scale Graph Attention Help Watershed Vegetation Forecasting?},
  author = {Ma, Andrew},
  booktitle = {9th ACM SIGSPATIAL International Workshop on AI for Geographic Information Systems (ACM GeoAI 2026)},
  year = {2026},
  url = {https://events.ornl.gov/acmsigspatial-geoai2026/event-agenda/},
  note = {Accepted to ACM SIGSPATIAL GeoAI 2026. Also submitted to NeurIPS TCCML Workshop. Solo author.}
}

AACE

Attention-Augmented Temporal Fusion Transformer for Short-Horizon Hypoglycemia Prediction in Type 1 Diabetes

Andrew Ma and Srihith Chennareddy

Endocrine Practice, 2026

Presented and published as first author at AACE / Endocrine Practice. Preprint on Research Square: https://www.researchsquare.com/article/rs-10856000/v1.

DOI Bib HTML

@article{ma2026attention,
  title = {Attention-Augmented Temporal Fusion Transformer for Short-Horizon Hypoglycemia Prediction in Type 1 Diabetes},
  author = {Ma, Andrew and Chennareddy, Srihith},
  journal = {Endocrine Practice},
  year = {2026},
  url = {https://www.endocrinepractice.org/article/S1530-891X(26)00841-4/abstract},
  doi = {10.21203/rs.3.rs-10856000/v1},
  note = {Presented and published as first author at AACE / Endocrine Practice. Preprint on Research Square: https://www.researchsquare.com/article/rs-10856000/v1.}
}

MLHMI / IEEE

CropViT: Climate-Aware Vision Transformer Model for Crop Yield Prediction

Srihith Chennareddy, Andrew Ma, and Yen-Chu Weng

In 2026 7th International Conference on Machine Learning and Human-Computer Interaction (MLHMI 2026), 2026

Accepted February 8, 2026. Published in IEEE Xplore (Doc ID: 11644961). First author oral presentation at JpGU-AGU Joint Meeting 2026 (Sun. May 24, 2026).

DOI Bib HTML

@inproceedings{chennareddy2026cropvit,
  title = {CropViT: Climate-Aware Vision Transformer Model for Crop Yield Prediction},
  author = {Chennareddy, Srihith and Ma, Andrew and Weng, Yen-Chu},
  booktitle = {2026 7th International Conference on Machine Learning and Human-Computer Interaction (MLHMI 2026)},
  year = {2026},
  doi = {10.23919/MLHMICPS00004.2026.00032},
  url = {https://ieeexplore.ieee.org/document/11644961},
  note = {Accepted February 8, 2026. Published in IEEE Xplore (Doc ID: 11644961). First author oral presentation at JpGU-AGU Joint Meeting 2026 (Sun. May 24, 2026).}
}

APS PNW

PSTL: Investigating Particle Scattering in Triply Periodic Minimal Surface Lattices

Andrew Ma, Srihith Chennareddy, C. Stoney, and 3 more authors

In Spring 2026 APS Northwest Section Meeting (APS PNW), 2026

Accepted May 21, 2026. First author presentation (Session N01.07).

Bib HTML

@inproceedings{ma2026pstl,
  title = {PSTL: Investigating Particle Scattering in Triply Periodic Minimal Surface Lattices},
  author = {Ma, Andrew and Chennareddy, Srihith and Stoney, C. and Brown, S. and Chennareddy, M. and Singh, A.},
  booktitle = {Spring 2026 APS Northwest Section Meeting (APS PNW)},
  year = {2026},
  url = {https://meetings-archive.aps.org/nws/2026/n01/7/},
  note = {Accepted May 21, 2026. First author presentation (Session N01.07).}
}

APS PNW

Quantifying Non-Gaussian Scattering Tails in Micro-Heterogeneous Phantoms for Proton Therapy Applications

Srihith Chennareddy, Andrew Ma, C. Stoney, and 2 more authors

In Spring 2026 APS Northwest Section Meeting (APS PNW), 2026

Accepted May 21, 2026. Presentation (Session P01.01).

Bib HTML

@inproceedings{chennareddy2026quantifying,
  title = {Quantifying Non-Gaussian Scattering Tails in Micro-Heterogeneous Phantoms for Proton Therapy Applications},
  author = {Chennareddy, Srihith and Ma, Andrew and Stoney, C. and Brown, S. and Singh, A.},
  booktitle = {Spring 2026 APS Northwest Section Meeting (APS PNW)},
  year = {2026},
  url = {https://meetings-archive.aps.org/nws/2026/p01/1/},
  note = {Accepted May 21, 2026. Presentation (Session P01.01).}
}

IEEE BIBM

CalAD: A Calibrated MRI Baseline for Alzheimer’s Detection on OASIS-3

Andrew Ma

In IEEE International Conference on Bioinformatics and Biomedicine (BIBM 2026) - ML4BMI Workshop, 2026

Accepted to IEEE BIBM ML4BMI Workshop 2026. Solo first author.

Bib

@inproceedings{ma2026calad_bibm,
  title = {CalAD: A Calibrated MRI Baseline for Alzheimer's Detection on OASIS-3},
  author = {Ma, Andrew},
  booktitle = {IEEE International Conference on Bioinformatics and Biomedicine (BIBM 2026) - ML4BMI Workshop},
  year = {2026},
  note = {Accepted to IEEE BIBM ML4BMI Workshop 2026. Solo first author.}
}

SPIE MI 2027

CalAD: A Calibrated MRI Baseline for Alzheimer’s Detection on OASIS-3

Andrew Ma and Khush Mehta

2027

Accepted as Poster at SPIE Medical Imaging 2027. First author (Ma, Andrew), second author (Mehta, Khush).

Bib

@article{ma2027calad_spie,
  title = {CalAD: A Calibrated MRI Baseline for Alzheimer's Detection on OASIS-3},
  author = {Ma, Andrew and Mehta, Khush},
  year = {2027},
  note = {Accepted as Poster at SPIE Medical Imaging 2027. First author (Ma, Andrew), second author (Mehta, Khush).}
}

Under Review

Spatial Deep Learning for Environmental DNA (eDNA) Localization and Biodiversity Mapping

Andrew Ma and Pedro Brandão Dias Ferreira Pinto

2026

Preprint / Under review. In collaboration with Pedro Brandão Dias Ferreira Pinto (University of Washington).

Bib

@article{ma2026edna,
  title = {Spatial Deep Learning for Environmental DNA (eDNA) Localization and Biodiversity Mapping},
  author = {Ma, Andrew and Pinto, Pedro Brand{\~a}o Dias Ferreira},
  year = {2026},
  note = {Preprint / Under review. In collaboration with Pedro Brand{\~a}o Dias Ferreira Pinto (University of Washington).}
}