TD Swinburne

Autonomous exploration at scale: TAMMBER
TAMMBER uses Bayesian methods to build kMC/Markov models with robust UQ on sampling incompleteness. We use this UQ to optimize resourse allocation in parallel computation, and bound predictions of diffusion models.
TDS† and D Perez, MSMSE (2022)
TDS† and D Perez, NPJ Computational Materials (2020) TAMMBER
TDS† and D Perez, PR Materials (2018)

ML-assisted QM/MM simulations
We use retrained ML potentials to embed DFT simulations in large-scale MD with seamless coupling. This permits DFT calculations at unprecedented scales.
P Grigorev et al., TDS†, Acta Materialia (2023) LML-retrain
TDS† and JR Kermode, PRB (2017)

Vibrational free energies beyond HTST: PAFI
PAFI is a path-constrained sampling routine, giving exact vibrational free energy differences for linear-scaling effort. This enables evaluation of activation free energies for very large systems which can not be treated by HTST approximations.
TDS† and M-C Marinica, PRL (2018) PAFI

Surrogate ML models for defect entropics
(PhD of Clovis Lapointe, with M-C Marinica)
Descriptor models typically used for energetics can predict the formation and migration entropy of defects. Allows a geometric interpretation of e.g. the Meyer-Neldel compensation law.
C Lapointe, TDS† et al. PRMaterials (2022)
C Lapointe, TDS† et al. PRMaterials (2020)

Analysis of Markov Models: PyGT
Much atomic data is a large network of metastable states. The Markov dynamics cannot be treated by usual routines (LAPACK) due to conditioning issues; PyGT provides a solution for these challenging networks.
TDS† and DJ Wales, JCTC (2020) PyGT

Mean-field models for lattice vibrations: BLaSa
Analytic mean-field approach for local anharmonic models, evaluating anharmonic vibrational phase free energies for milliseconds of effort. Current developments are within the PhD of R Dsouza since 2021.
TDS† et al., J Neugebauer PRB Rapid Communications (2020) BLaSa