Geometric Deep Learning

Neural architectures that encode symmetry, geometry, equivariance, and three-dimensional structure.

  1. How invariant attention scores, equivariant values, and energy-based force prediction turn geometric Transformers into practical interatomic potentials.
    Tutorial · Published Aug 08, 2026 · Updated Aug 09, 2026 · 38 min read
  2. How canonicalization, local frames, frame averaging, and probabilistic symmetrization create geometric models—and why continuity is difficult.
    Tutorial · Published Aug 08, 2026 · Updated Aug 09, 2026 · 37 min read
  3. Flow matching beyond Euclidean space, from tangent velocity fields and geodesic conditional paths to product manifolds for molecular geometry.
    Tutorial · Published Aug 08, 2026 · Updated Aug 09, 2026 · 33 min read
  4. How geometric graph networks move from invariant distances and angles to equivariant coordinates and vector channels—and what directionality buys.
    Tutorial · Published Aug 08, 2026 · Updated Aug 09, 2026 · 35 min read
  5. How irreducible rotation types, spherical harmonics, and Clebsch–Gordan tensor products create expressive equivariant neural-network layers.
    Tutorial · Published Aug 08, 2026 · Updated Aug 09, 2026 · 33 min read
  6. Understanding the spherical equivariant layers that power modern molecular neural networks, from group theory foundations to Clebsch-Gordan tensor products.
    Tutorial · Published Feb 02, 2026 · Updated Aug 09, 2026 · 37 min read