notesum.ai

Published at October 18

Almost-Linear RNNs Yield Highly Interpretable Symbolic Codes in Dynamical Systems Reconstruction

cs.CE

cs.LG

Released Date: October 18, 2024

Authors: Manuel Brenner¹, Christoph Jürgen Hemmer¹, Zahra Monfared², Daniel Durstewitz¹

Aff.: ¹Dept. of Theoretical Neuroscience, Central Institute of Mental Health, Medical Faculty, Heidelberg University, Germany; ²Interdisciplinary Center for Scientific Computing (IWR), Heidelberg University, Germany

Arxiv: https://arxiv.org/abs/2410.14240v1