Unsupervised learning of stationary and switching dynamical system models from Poisson observations
Christian Y Song, Maryam M Shanechi · 2023
Read Unsupervised learning of stationary and switching dynamical system models from Poisson observations by Christian Y Song, Maryam M Shanechi (2023) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-10714100.txt.
Description
Christian Song and Maryam Shanechi present in a 2023 paper a new way to learn brain activity models from spike data with no labels. Spike counts follow a Poisson pattern, and models that switch between simple regimes can explain rich behavior. Without true regime labels, learning must be unsupervised, which is where past methods slip.
Their filter samples points by fixed rules from a grid method rather than by a rough curve fit. That choice tracks the true spread of hidden states more faithfully in tests. Set inside a learning loop, it taught both steady and switching models with high accuracy. Brain and fake data both back the gain over past filters. The work aids tools that read neural dynamics for science and care.
Catalog record
| CREATOR | Christian Y Song, Maryam M Shanechi |
|---|---|
| YEAR | 2023 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 196 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |