Skip to content

Dynamical Symmetric Information Bottleneck (DySIB)

Home / Concepts / Dynamical Symmetric Information Bottleneck (DySIB)

Dynamical Symmetric Information Bottleneck (DySIB)

Auto-generated stub. Edit this file to add more details.

A method for learning low-dimensional representations of time-series by maximizing predictive mutual information between temporal windows while penalizing complexity.

Why It Matters

The central methodological innovation of the paper for latent dynamics discovery without reconstruction, providing a reusable framework for time-series representation learning.

Evidence

Here we introduce DySIB (Dynamical Symmetric Information Bottleneck) as a method to learn low-dimensional representations of time-series data by maximizing predictive mutual information between past and future observation windows while penalizing representation complexity.

Metadata & Links