Sampling-Optimisation-Geometry-Bench
GeometryBench
Manifolds, couplings and transport.
Public-dev · shared task identities · Dataset manifest
Choose a task
Wasserstein symmetry without an optimal plansogb-004 · NL L3→Why a sphere gradient is a projected ambient gradientsogb-005 · NL L3→Which space contains the sphere gradient?sogb-011 · NL L1→What information does a coupling contain?sogb-012 · NL L1→Recovering joint probability mass from one marginalsogb-015 · NL L2 · Lean L2→A quadratic gradient through the actual sphere differentialsogb-016 · NL L3 · Lean L3→Gluing the right joint law before applying Minkowskisogb-018 · NL L4 · Lean L4→From a selected-curve chord inequality to its first-order boundsogb-022 · NL L3 · Lean L3→Scope and difficulty
These are views of the same catalog, not new datasets. A cross-listed core keeps its identity, family, split and source binding. NL/Lean levels are provisional, not empirically calibrated.
Topics: Riemannian Optimisation, Statistical Optimal Transport.