Time-lagged autoencoders: Deep learning of slow collective variables for molecular kinetics

2017/10/30 by Christoph Wehmeyer, Frank Noé · 1 citation
Biochemistry, Genetics and Molecular Biology · Computer Science · Materials Science · Mathematics · Physics and Astronomy · Psychology · #Artificial intelligence #Computer science #Deep learning #Econometrics #Kinetics #Machine Learning in Materials Science #Mathematics #Model Reduction and Neural Networks #Physics #Protein Structure and Dynamics #Psychology #cs.LG #physics.bio-ph #physics.chem-ph #stat.ML

paper · pdf · doi:10.1063/1.5011399

arxiv created 2017/10/30 · openalex publication_date 2018/03/15 · arxiv updated 2018/04/04 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06

Abstract

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.

Citations

• Adam: A Method for Stochastic Optimization
• A Data-Driven Approximation of the Koopman Operator: Extending Dynamic Mode Decomposition
• Screen Savers of the World Unite!
• Matplotlib: A 2D Graphics Environment
• RELATIONS BETWEEN TWO SETS OF VARIATES
• Spectral Properties of Dynamical Systems, Model Reduction and Decompositions
• Atomic-Level Characterization of the Structural Dynamics of Proteins
• Markov models of molecular kinetics: Generation and validation
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Mastering the game of Go with deep neural networks and tree search
• Quantum-chemical insights from deep tensor neural networks
• Automatic chemical design using a data-driven continuous representation of molecules
• A variational conformational dynamics approach to the selection of collective variables in metadynamics
• Variational approach for learning Markov processes from time series data
• Deep learning

Cited by

• Variational encoding of complex dynamics
• Exploring the Potential of Parallel Biasing in Flat Histogram Methods