PhD Student @ UW CSE

Mike Merrill

About Me

I am a fifth year PhD student at the Paul G. Allen School of Computer Science & Engineering at The University of Washington, where I am advised by Tim Althoff. I am affiliated with UW NLP.

I am looking for a full time job! If you think we would be a good match, please reach out. Previously, I was a Student Researcher at Google Research, an ML Research Intern at Apple Health AI and a data scientist and the second full-time employee at HealthRhythms.

Research Interests

My work focuses on methods, datasets, and benchmarks for training and evaluating large models (including language models) on time series data and code. Recently I have been teaching multimodal language models to reason about time series, and building LLM agents that can make decisions and take actions based off of these inputs.

Publications

Are Language Models Actually Useful for Time Series Forecasting?

Mingtian Tan, Mike A. Merrill, Vinayak Gupta, Tim Althoff, and Thomas Hartvigsen

Preprint, 2024 [PDF]

Transforming Wearable Data into Health Insights using Large Language Model Agents

Mike A. Merrill, Akshay Paruchuri, Naghmeh Rezaei, Geza Kovacs, Javier Perez, Yun Liu, Erik Schenck, Nova Hammerquist, Jake Sunshine, Shyam Tailor, Kumar Ayush, Hao-Wei Su, Qian He, Cory Y. McLean, Mark Malhotra, Shwetak Patel, Jiening Zhan, Tim Althoff, Daniel McDuff, and Xin Liu

Preprint, 2024 [PDF] [Google Research Blog]

Language Models Still Struggle to Zero-shot Reason about Time Series

Mike A. Merrill, Mingtian Tan, Vinayak Gupta, Tom Hartvigsen, and Tim Althoff

Preprint, 2024 [PDF] [Data & Code]

Homekit2020: A Benchmark for Time Series Classification on a Large Mobile Sensing Dataset with Laboratory Tested Ground Truth of Influenza Infections

Mike A. Merrill, Esteban Safranchik, Arinbjorn Kolbeinsson, Piyusha Gade, Ernesto Ramirez, Ludwig Schmidt, Luca Foshchini, and Tim Althoff

CHIL, 2023 [PDF]

Self-supervised Pretraining and Transfer Learning Enable Flu and COVID-19 Predictions in Small Mobile Sensing Datasets

Mike A. Merrill and Tim Althoff

CHIL, 2023 [PDF]

CORAL: COde RepresentAtion Learning with Weakly-Supervised Transformers for Analyzing Data Analysis

Ge Zhang, Mike A. Merrill, Yang Liu, Jeffrey Heer, and Tim Althoff

EPJ Data Science, 2022 [PDF] *Co-First Author

Globem dataset: Multi-year datasets for longitudinal human behavior modeling generalization

Xuhai Xu, Han Zhang, Yasaman Sefidgar, Yiyi Ren, Xin Liu, Woosuk Seo, Jennifer Brown, Kevin Kuehn, Mike A. Merrill, Paula Nurius, Shwetak Patel, Tim Althoff, Margaret E. Morris, Eve Riskin, Jennifer Mankoff, and Anind K. Dey

NeurIPS, 2022 [PDF]

MULTIVERSE: Mining Collective Data Science Knowledge from Code on the Web to Suggest Alternative Analysis Approaches

Mike A. Merrill, Ge Zhang, and Tim Althoff

KDD, 2021 [PDF]

CrossCheck: Integrating self-report, behavioral sensing, and smartphone use to identify digital indicators of psychotic relapse

Dror Ben-Zeev, Rachel Brian, Rui Wang, Weichen Wang, Andrew T. Campbell, Min S. H. Aung, Michael Merrill, Vincent W. S. Tseng, Tanzeem Choudhury, Marta Hauser, John M. Kane, and Emily A. Scherer

Psychiatric Rehabilitation Journal, 2017 [PDF]

CrossCheck: toward passive sensing and detection of mental health changes in people with schizophrenia

Rui Wang, Min S. H. Aung, Saeed Abdullah, Rachel Brian, Andrew T. Campbell, Tanzeem Choudhury, Marta Hauser, John Kane, Michael Merrill, Emily A. Scherer, Vincent W. S. Tseng, and Dror Ben-Zeev

Ubicomp, 2016 [PDF]

Assessing mental health issues on college campuses: Preliminary findings from a pilot study

Vincent W. S. Tseng, Michael Merrill, Franziska Wittleder, Saeed Abdullah, Min Hane Aung, and Tanzeem Choudhury

Ubicomp, 2016 [PDF]

Posters

Sensing depression: Using smartphone sensors to predict changes in depression severity

Ellen Frank, Michael Merrill, Hane Aung, Claudio Soares, Sidney Kennedy, Mark Matthews, David Kupfer, Aleksandra Lalovic, and Tanzeem Choudhury

The 56th Annual Meeting of the American College of Neuropsychopharmacology, 2017

Continuous behavioral data as a potential depression biomarker