AI Paper Atlas

Research map / Probabilistic and scientific modeling / Dynamical systems

Decomposed Linear Dynamical Systems (dLDS) for learning the latent components of neural dynamics

Noga Mudrik, Yenho Chen, Eva Yezerets, Christopher J. Rozell, Adam S. Charles

JMLR 2024 · Probabilistic and scientific modeling › Dynamical systems · approach: Switching and law-discovery dynamics · 8 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace discrete switching states with sparse continuous coefficients (dLDS vs rSLDS)linear dynamical system model · Synthetic speed (spiral with four speeds)representation efficiency: one dictionary element vs four non-zero discrete statesImproved
replace discrete switching states with continuous coefficientsrSLDS · Synthetic rotation on spherenumber of operators needed: two DOs vs four non-zero discrete statesImproved
use dLDS instead of rSLDS with one state (M=1) for stability shiftlinear dynamical system · decaying then growing spiralreconstruction of stability shiftImproved
use dLDS instead of rSLDS with two states for smooth transitionslinear dynamical system · smooth sigmoid transition between two rotationsreconstruction of dynamicsImproved
use dLDS instead of rSLDS for two unsynchronized subpopulationslinear dynamical system · 10-dim, 6 ground truth DOs, 50 draws, M=15operator recovery correlation: correlation approx 1Improved
add L1 sparsity regularization on dynamics coefficients (ablation)dLDS · FitzHugh-Nagumocoefficient interpretability: R2 1.0 -> 1.0No clear effect
add L1 sparsity regularization on dynamics coefficients (ablation)dLDS · Lorenz attractorone-step prediction R2: 1.0 -> 0.70Hurt
use dLDS instead of rSLDSone-step prediction · C. elegans calcium imagingR2: 0.92 -> 0.74Hurt
use dLDS instead of rSLDSone-step prediction · Unregularized LorenzR2: 0.99 -> 1.0Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy