Understanding Generalization and Robustess of Learned Representations of Chaotic Dynamical Systems

Luã Streit, Vikram Voleti, Tegan Maharaj

Keywords: Representation Learning, Deep learning, Dynamical Systems, Generalization, Robustness, Time Series, Ordinary Differential Equations, OOD Generalization

TL;DR: We propose an evaluation framework to study the generalization of learned representations and apply it to chaotic dynamical systems.

Abstract:

We investigate the generalization capabilities of different methods of learning representations via an extensible synthetic dataset of real-world chaotic dynamical systems introduced by Gilpin (2021).
We propose an evaluation framework built on top of this dataset, called ValiDyna, which uses probes and multi-task learning to study robustness and out-of-distribution (OOD) generalization of learned representations across a range of settings, including changes in losses, architecture, etc. as well as changes in the distribution of the dynamical systems' initial conditions and parameters.
Our evaluation framework is of interest for generalization and robustess broadly, but we focus our assessment here on evaluating learned representations of ecosystem dynamics, with the goal of using these representations in ecological impact assesments, with applications to biodiversity conservation and climate change mitigation.

Confirmation: Yes