I am a 3rd year PhD candidate in Computational Materials Science and Engineering at MIT. I work on differentiable simulations, generative models, and generally deep learning for accelerated discovery of materials (polymers, molecules). I'm advised by Rafael Gomez-Bombarelli.
End-to-end learning of unknown physical terms in partial differential equations with differentiable PDE solvers (finite element or spectral methods). Here, by applying PDE-constrained optimization, we can learn cure kinetics terms in frontal polymerization processes.
Ongoing work: Learning closure terms in the PDE from multimodal experimental data (thermal capture videos and calorimetry curves).
Monte Carlo Simulations of Thermoset Polymer Chain Growth and Crosslinking
Reverse Monte Carlo method to model the complex graph network of degradable crosslinked copolymers. We can now retrospectively determine network structure and reaction parameters by matching with experimental fragment spectra for the first time.