Atomic architect. Wrangler of neural nets. Geometry junkie. blondegeek.

Resume / CV

Past Talks

Mischief

Upcoming Travel and (Remote) Talks

• AI/ML for ALS Synchrotron Science Innovation forum, Berkeley Lab, August 11, 2020.
• Lecture for the Berkeley Lab Deep Learning School, Berkeley Lab, September 3, 2020.
• Seminar at FHI Berlin, September 23, 2020.
• Keynote for Applied Computational Sciences symposium (ACOS), October 28, 2020.

Code

• e3nn_tutorial: Tutorial on Euclidean symmetry equivariant neural networks

  This tutorial uses a combination of slides and interactive jupyter notebooks using the e3nn framework to present fundamental concepts about Euclidean equivariant neural networks: data types, equivariant operations, how symmetry effects simple tasks, and how to construct the building blocks of e3nn. This tutorial was originally presented at IPAM at UCLA on November 14, 2019.

• e3nn: A framework for Euclidean symmetry equivariant neural networks

  Convolutional neural networks for point clouds and voxels with Euclidean symmetries (3D translation and 3D rotation). The current code is a merger of the work from Tensor field networks and 3D Steerable CNNs.

Recent pre-prints and publications

• Finding Symmetry Breaking Order Parameters with Euclidean Neural Networks

  Tess E. Smidt, Mario Geiger, and Benjamin Kurt Miller (arXiv:2007.02005)
  
  code: (notebook / html)

• Anisotropic 2D Excitons Unveiled in Organic-Inorganic Quantum Wells

  Maserati, L. et al, Materials Horizons, 2020, DOI: 10.1039/C9MH01917K

• An Automatically Curated First-Principles Database of Ferroelectrics

  Smidt, T.E., Mack, S.A., Reyes-Lillo, S.E. et al. Sci Data 7, 72 (2020).
  
  code:
  pymatgen.analysis.ferroelectricity /
  atomate…workflow…ferroelectric
  
  data:
  ferroelectric_search_site
  figshare data deposit

• Tensor Field Networks

  code: paper repo / recent – e3nn