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.

Second-Order Methods in JAX

Adaptive Circular Manifold (ACM): A Density-Dependent Geometric Metric for Robust Gradient Descent

Information geometry on the probability simplex

InfoGeoLab — Interactive Information Geometry & Statistical Manifold Laboratory. 6 modules: manifold explorer, Fisher info, geodesics, natural gradient, divergences, exponential families.

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

This project simulates a Bayesian adaptive Phase II clinical trial evaluating a hypothetical therapy that modulates circulating IGF-1 (insulin-like growth factor 1) in colorectal cancer patients. The primary outcome is the continuous change in log-transformed IGF-1 from baseline to 12 weeks.