Learning Clustered Representation for Complex Free Energy Landscapes

Abstract

In this paper we first analyzed the inductive bias underlying the data scattered across complex free energy landscapes (FEL), and exploited it to train deep neural networks which yield reduced and clustered representation for the FEL. Our parametric method, called Information Distilling of Metastability (IDM), is end-to-end differentiable thus scalable to ultra-large dataset. IDM is also a clustering algorithm and is able to cluster the samples in the meantime of reducing the dimensions. Besides, as an unsupervised learning method, IDM differs from many existing dimensionality reduction and clustering methods in that it neither requires a cherry-picked distance metric nor the ground-true number of clusters, and that it can be used to unroll and zoom-in the hierarchical FEL with respect to different timescales. Through multiple experiments, we show that IDM can achieve physically meaningful representations which partition the FEL into well-defined metastable states hence are amenable for downstream tasks such as mechanism analysis and kinetic modeling.

6 figures, 1 table in the main text

References in corpus (12)

• Adam: A Method for Stochastic Optimization
• Escaping free-energy minima
• An Overview of Multi-Task Learning in Deep Neural Networks
• SchNet - a deep learning architecture for molecules and materials
• Identification of slow molecular order parameters for Markov model construction
• Adversarially Learned Inference
• Diffusion Maps, Spectral Clustering and Eigenfunctions of Fokker-Planck operators
• Thinking Fast and Slow with Deep Learning and Tree Search
• Recognizing molecular patterns by machine learning: an agnostic structural definition of the hydrogen bond
• Recognizing Local and Global Structural Motifs at the Atomic Scale
• Invariant Information Clustering for Unsupervised Image Classification and Segmentation
• Fast Adaptation in Generative Models with Generative Matching Networks

Cited by in corpus (2)

• A Perspective on Deep Learning for Molecular Modeling and Simulations
• Deep Reinforcement Learning of Transition States