Machine Learning for Phase Behavior in Active Matter Systems
This work provides an effective computational method for determining the phase behavior of active matter systems, which is useful for researchers studying complex phase diagrams in physics and materials science.
This paper uses deep learning to predict motility induced phase separation (MIPS) in active Brownian particle (ABP) suspensions by classifying individual particles into phases. The method accurately determines the fraction of dilute particles, allowing the system's state (homogeneous dilute, dense, or coexistence) to be identified, showing strong agreement with MIPS binodal simulations.
We demonstrate that deep learning techniques can be used to predict motility induced phase separation (MIPS) in suspensions of active Brownian particles (ABPs) by creating a notion of phase at the particle level. Using a fully connected network in conjunction with a graph neural network we use individual particle features to predict to which phase a particle belongs. From this, we are able to compute the fraction of dilute particles to determine if the system is in the homogeneous dilute, dense, or coexistence region. Our predictions are compared against the MIPS binodal computed from simulation. The strong agreement between the two suggests that machine learning provides an effective way to determine the phase behavior of ABPs and could prove useful for determining more complex phase diagrams.