Summary of On Learning What to Learn: Heterogeneous Observations Of Dynamics and Establishing (possibly Causal) Relations Among Them, by David W. Sroczynski et al.
On Learning what to Learn: heterogeneous observations of dynamics and establishing (possibly causal) relations among them
by David W. Sroczynski, Felix Dietrich, Eleni D. Koronaki, Ronen Talmon, Ronald R. Coifman, Erik Bollt, Ioannis G. Kevrekidis
First submitted to arxiv on: 10 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Dynamical Systems (math.DS); Numerical Analysis (math.NA)
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 The proposed method demonstrates two data-driven approaches to determine the relevant quantities between multiple heterogeneous time series from a physical system. By processing ensembles of time series, the approach identifies common observables and information particular to each process. This allows any function approximation technique, such as k-nearest neighbors, Geometric Harmonics, Gaussian Processes, or Neural Networks, to be used to learn the input-output relation. The method also explores two twists: identifying quantities of interest from measurements and relating the framework to causality by modeling system evolution. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper shows how to use many different types of data to figure out what’s important in a physical process. It does this by looking at lots of time series data from different places and times. The approach helps identify what’s common between these data streams and what’s unique to each one. Then, any machine learning method can be used to learn how the different parts are related. |
Keywords
» Artificial intelligence » Machine learning » Time series
