Constructing Custom Thermodynamics Using Deep Learning
This provides a general methodology for automated scientific discovery in complex phenomena, such as polymer dynamics, where traditional intuition may fail.
The researchers tackled the problem of learning macroscopic dynamical descriptions of stochastic dissipative systems from microscopic data, using a deep learning platform based on the Onsager principle, and demonstrated it by constructing three interpretable thermodynamic coordinates and a dynamical landscape for polymer stretching, including stable and transition states and control of stretching rates.
One of the most exciting applications of artificial intelligence (AI) is automated scientific discovery based on previously amassed data, coupled with restrictions provided by known physical principles, including symmetries and conservation laws. Such automated hypothesis creation and verification can assist scientists in studying complex phenomena, where traditional physical intuition may fail. Here we develop a platform based on a generalized Onsager principle to learn macroscopic dynamical descriptions of arbitrary stochastic dissipative systems directly from observations of their microscopic trajectories. Our method simultaneously constructs reduced thermodynamic coordinates and interprets the dynamics on these coordinates. We demonstrate its effectiveness by studying theoretically and validating experimentally the stretching of long polymer chains in an externally applied field. Specifically, we learn three interpretable thermodynamic coordinates and build a dynamical landscape of polymer stretching, including the identification of stable and transition states and the control of the stretching rate. Our general methodology can be used to address a wide range of scientific and technological applications.