Time-lagged autoencoders: Deep learning of slow collective variables for molecular kinetics
This addresses the challenge of analyzing high-dimensional molecular kinetics data for researchers in physical and chemical sciences, representing an incremental improvement over existing methods.
The authors tackled the problem of reducing the dimensionality of molecular dynamics data by developing a time-lagged autoencoder, which reliably finds low-dimensional embeddings that capture slow dynamics, outperforming linear techniques.
Inspired by the success of deep learning techniques in the physical and chemical sciences, we apply a modification of an autoencoder type deep neural network to the task of dimension reduction of molecular dynamics data. We can show that our time-lagged autoencoder reliably finds low-dimensional embeddings for high-dimensional feature spaces which capture the slow dynamics of the underlying stochastic processes - beyond the capabilities of linear dimension reduction techniques.