I’m a Ph.D Candidate in the Data Science program at Worcester Polytechnic Institute studying sequential representation learning, advised by Prof. Elke Rundensteiner and Prof. Xiangnan Kong and collaborating with Cansu Sen in the Data Science Research Group.
My high-level interest is in mental models and their role in Artificial Intelligence, especially in planning problems. I am particularly interested in how we learn and curate our mental models and would eventually like to approach this representation learning problem through the lense of early childhood development. During my PhD so far, my research has involved the combination of Recurrent Neural Networks and some Reinforcement Learning techniques (which together loosely emulate human decision making). I have so far applied these techniques on modeling clinical decision making scenarios using both clinical time series and sequences of notes, studying irregular time series, early classification, attributed document classification, and medical word embedding.
Please feel free to contact me with questions regarding my research or Data Science at WPI at email@example.com.
- [KDD] Thomas Hartvigsen, Cansu Sen, Xiangnan Kong, Elke Rundensteiner. Adaptive-Halting Policy Network for Early Classification. ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2019. pdf.
- [BHI] Jidapa Thadajarassiri, Cansu Sen, Thomas Hartvigsen, Xiangnan Kong, Elke Rundensteiner. Comparing General and Locally-Learned Word Embeddings for Clinical Text Mining. IEEE International Conference on Biomedical and Health Informatics (BHI), 2019.
- [CCIS] Hartvigsen, T., Sen, C., Rundensteiner, E. Detecting MRSA Infections by Fusing Structured and Unstructured Electronic Health Record Data. To appear in Communications in Computer and Information Science.
- [HEALTHINF] Hartvigsen, T., Sen, C., Brownell, S., Teeple, E., Kong, X. and Rundensteiner, E. (2018). Early Prediction of MRSA Infections using Electronic Health Records. International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2018) - Volume 5: HEALTHINF, pages 156-167, ISBN: 978-989-758-281-3. Shortlisted for Best Student Paper. pdf.
- [ECML/PKDD] Sen, C., Hartvigsen, T., Claypool, K., Rundensteiner, E. (2017). CREST - Risk Prediction for Clostridium Difficile Infection Using Multimodal Data Mining. ECML/PKDD 2017. pdf.
In my spare time, I enjoy rock climbing, cycling, reading (fantasy, science fiction, science fact), and playing guitar.
- May 10, 2019: Poster Presentation - Presented a poster titled “Adaptive-Halting Policy Network for Early Classification” at the New England Machine Learning Day at Northeastern University.
- April 29, 2019: Paper Accepted to ACM SIGKDD - Our paper titled “Adaptive-Halting Policy Network for Early Classification” was accepted to KDD 2019, research track, acceptance 14%.
- April 16, 2019: Passed Research Qualifier - Passed my research qualifier, studying the classification of irregular multivariate time series.
- April 9, 2019: First place in poster competition - I was awarded first place in WPI’s annual graduate research poster competition.
- March 5, 2019: Paper Accepted to BHI - Our paper titled “Comparing General and Locally-Learned Word Embeddings for Clinical Text Mining” was accepted to the IEEE International Conference on Biomedical and Health Informatics.
- January 18, 2019: Workshop - Attended the Geometric Analysis Approach to AI Workshop at Harvard University.
- November 15, 2018: Defended Master’s Thesis - Successfully defended Master’s Thesis: Adaptively-Halting RNN for Tunable Early Classification of Time Series.
- September 1, 2018: Machine Learning Internship - Beginning a part-time internship with UMass Medical School in the quantitative sciences department.
- August 02, 2018: REU Project Complete - Mentored two undergraduate REU students, resulting in one paper and one poster. They submitted a paper to the MIT URTC and their poster will be presented at the Research Experience for Undergraduates Symposium in Washington, D.C.
- May 30, 2018: Mentoring REU students - I am working with 2 undergraduate Research Experience for Undergraduates (REU) students this summer to develop a state-of-the-art model for classifying time series with missing values.
- April 30, 2018: AWARD - I was awarded the WPI Data Science Citizen Award for involvement and participation in the growth of the department.
- April 24, 2018: Poster Presentation - Poster presentation at the WPI Graduate Research and Innovation Exchange. We had three posters at this event, one of which won the People’s Choice Award.
- January 19, 2018: Conference Talk - I presented our paper “Early Prediction of MRSA Infections Using Electronic Health Records” at HEALTHINF 2018 in Madeira, Funchal.
- December 12, 2017: Paper award - Our paper, “Early Prediction of MRSA Infections Using Electronic Health Records”, was nominated for the HEALTHINF 2018 best student paper award.
- October 16, 2017: Paper Accepted to HEALTHINF - Our paper, “Early Prediction of MRSA Infections Using Electronic Health Records”, was accepted to HEALTHINF 2018 as a full paper.
- August 19, 2017: Conference Attendence - Attended Machine Learning for Healthcare (MLHC) 2017 at Northeastern University.
- June 23, 2017: Paper Accepted to ECML - Our paper, “CREST - Risk Prediction for Clostridium Difficile Infection Using Multimodal Data Mining”, was accepted to ECML-PKDD 2017 as a full paper.
- May 2, 2017: AWARD - I was awarded the WPI Data Science Citizen Award for involvement and participation in the growth of the department.
- February 8, 2017: AWARD - Poster presentation with Cansu Sen at the WPI Graduate Research and Innovation Exchange, received the People’s Choice award.
- August 25, 2016 - Began Ph.D. at Worcester Polytechnic Institute, Worcester, Massachusetts.