Software
Recent Papers:
JaxSGMC: Modular stochastic gradient MCMC in JAX
Published code:
Partial charge prediction trained with AL using Dropout MC
Code for partial charge prediction graph neural network trained with active learning using Dropout Monte Carlo.
ML potentials trained on DFT and EXP data
Code for training ML potentials concurrently on ab initio and experimental data.
Relative Entropy Minimization
Code for training of coarse-grained molecular models with relative entropy minimization.
Jax / Haiku implementation of DimeNet++
This DimeNet++ implementation can be used for molecular dynamics simulations with Jax, M.D. and for molecular property prediction.