Sarah Marzen
  • Main
  • People
  • Contact
  • Google Scholar
  • Random ruminations
  • Research Program
  • SFI Workshop: Sensory Prediction, Engineered and Evolved
  • Main
  • People
  • Contact
  • Google Scholar
  • Random ruminations
  • Research Program
  • SFI Workshop: Sensory Prediction, Engineered and Evolved

Random musings

The problem with not knowing dynamical systems

2/10/2021

0 Comments

 
I don't usually like to write about stuff like this, but I feel like this might actually do some good.  Myself and Jim Crutchfield (well-known for his work on chaos theory) have papers about new methods for continuous-time, discrete-event process inference and prediction (here) and about how one can view the predictive capabilities of dynamical systems as a function of their attractor type (here).  The reviews-- one from an information theory journal and another from machine learning experts-- unfortunately illustrated a lack of common knowledge on interdisciplinary problems.  So I thought I'd put a few key points here, for those studying recurrent neural networks in any way, shape, or form.

First, if you have a dynamical system, you can classify its behavior qualitatively by attractor type.  There are three types of attractors: fixed points, limit cycles, and beautiful strange attractors.  It turns out that the "qualitative" attractor type is a guide to many computational properties of the dynamical system (again, soon to appear on arXiv).

Second, hidden Markov models-- including unifilar ones, in which the current state and next symbol determine the next state-- are not memoryless or Markovian.

More to come.
0 Comments



Leave a Reply.

    Author

    Write something about yourself. No need to be fancy, just an overview.

    Archives

    November 2022
    July 2022
    December 2021
    November 2021
    March 2021
    February 2021
    May 2020
    July 2019
    May 2017

    Categories

    All

    RSS Feed

Proudly powered by Weebly