Summary of Discovering Group Dynamics in Coordinated Time Series Via Hierarchical Recurrent Switching-state Models, by Michael T. Wojnowicz et al.
Discovering group dynamics in coordinated time series via hierarchical recurrent switching-state models
by Michael T. Wojnowicz, Kaitlin Gili, Preetish Rath, Eric Miller, Jeffrey Miller, Clifford Hancock, Meghan O’Donovan, Seth Elkin-Frankston, Tad T. Brunyé, Michael C. Hughes
First submitted to arxiv on: 26 Jan 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary As machine learning educators, we present a novel hierarchical switching-state model for modeling and forecasting the behavior of multiple interacting agents. This unsupervised approach learns both system-level and individual-level dynamics simultaneously, incorporating top-down influence from system-level Markov chains to entity-level latent chains that govern observed time series. Our recurrent feedback mechanism allows recent situational context to inform dynamics at all levels, improving interpretability and reducing error in forecasting. We demonstrate competitive performance compared to larger neural networks on various datasets, including synthetic basketball team movements and soldier data. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary We’re working on a new way to understand how groups of things move together over time. Right now, most models focus on individual actions, but we want to include the bigger picture – what’s happening in the group as a whole? Our approach uses a special kind of computer program that can learn from many different data sets and figure out both what the individual entities are doing and how they’re working together. This helps us make better predictions about what will happen next, and it gives us useful insights into group dynamics. |
Keywords
* Artificial intelligence * Machine learning * Time series * Unsupervised
