Tom Hartvigsen
Thomas (Tom) Hartvigsen
Postdoctoral Associate
MIT, CSAIL
Cambridge, MA
Pronouns: He/Him
What’s New?
ICDM'22 paper on explainability for deep time series classifiers
CIKM'22 paper on early classification of irregular time series
CIKM'22 paper on recurrent classifier chains with missing labels
Co-chairing NeurIPS workshop
Invited talk at University of Rochester
Invited talk at Tufts
Invited talk at Google Research
ACL'22 paper on constructing datasets with language models.
Invited talk at MIT Horizons
AAAI'22 paper on positive unlabeled learning.
SDM'22 paper on positive unlabeled learning.
About
I am a Postdoctoral Associate at MIT in the Computer Science and Artificial Intelligence Laboratory (CSAIL) with Prof. Marzyeh Ghassemi. My interests span Time Series, Machine Learning, Data Mining, NLP, Fairness in AI Systems, and applications to understand and improve Health.
Research
I build machine learning systems from partial information. The projects that excite me the most: (1) robustly model complex time-varying systems, (2) prevent perpetuating bias via machine learning, and/or (3) have impact through real-world deployment.
Some recent highlights:
TOXIGEN: constructing large, diverse hate speech datasets with large language models to train better hate speech classifiers (ACL'22)(dataset)
Learning to stop early and classify ongoing time series in time-sensitive domains (see KDD'19, KDD'20,CIKM'22 papers)
Methods for recovering models of annotators' partial sequential labeling behavior from machine learning datasets (see AAAI'22, SDM'22, CIKM'22 papers)
Explainability for time series and NLP models (see CIKM'21, ACL'20, and ICDM'22 papers)
All papers are collaborations, please see my full list of publications.
Service
Organizer: NeurIPS 2022 Workshop on Learning from Time Series for Health; CHIL 2023.
Program Committee: AAAI, CVPR, ICCV, ACL, EMNLP, NAACL, KDD, NeurIPS Datasets & Benchmarks Track
Workshop Reviewer: epiDAMIK @ KDD 2022, RobustSeq @ NeurIPS 2022
Led tutorial on Deep Learning with PyTorch for undergrads
WPI data science graduate student council, WPI graduate student senate, WPI Arts & Sciences council, Organized WPI deep learning reading group
In the News
Our work on TOXIGEN was covered by TechCrunch and Microsoft Research
Our work on Fair Explainability was covered by MIT News
History
Before MIT, I received my PhD in Data Science from Worcester Polytechnic Institute in 2021, where I was a GAANN Fellow advised by Professors Elke Rundensteiner and Xiangnan Kong. I also interned at Microsoft and UMass Medical School. Before graduate school, I received a B.A. in Applied Math from SUNY Geneseo in 2016.
Misc
Outside of research, I enjoy bouldering, biking, books (science fiction/science fact), birding, juggling, vegan cooking, and guitar. I also spent a summer living at BioSphere 2 in rural Arizona.