self-attention neural torch ansatz for electronic structure calculations introduced in the paper “A Self-attention Ansatz for ab initio Quantum Chemistry” from Google Deep Mind.

Stochastic Second-Order Methods in JAX

Tungsten Alpha: A JAX-native, 7-layer Adelic-Riemannian Operating System. Implements resurgent gradient flow through discrete logic and Levi-Civita parallel transport for decentralized XLA clusters. Hardened for zero-jitter, systolic array execution.

Six papers on information geometry, natural gradient methods, and value-perturbation theory with full experimental code

Information geometry on the probability simplex