Selected highlights
A curated set of posts I recommend as starting points.
- MADField learns adsorbate density fields from cDFT and GCMC, enabling fast uptake prediction and database-scale methane screening.
- How HACO, a Human–AI Co-discovery system, produced MaskGXT, a competitive generative model for crystal structure prediction.
- Statistical mechanics: from Newton's equations to ensembles, thermostats, barostats, Monte Carlo, and connections to generative modeling.
- GFlowNets from a probabilistic-ML perspective: reward-proportional sampling, training objectives, and connections to MaxEnt RL and variational inference.
- How path measures connect Jarzynski's equality, free-energy estimation, annealed importance sampling, diffusion models, and GFlowNets.
- Three routes to the Fokker-Planck equation—physical intuition, heuristic discretization, and a rigorous derivation with Itô calculus.
- Quantum chemistry and density functional theory: from the Schrödinger equation to Kohn-Sham DFT and modern deep learning approaches.
- Understanding the spherical equivariant layers that power modern molecular neural networks, from group theory foundations to Clebsch-Gordan tensor products.