Establishes W1 quantitative bounds for Laplace-type convergence of measures with norm-like potentials using coarea formula under generalized Jacobian invertibility, applied to maxent and SGLD.
Bayesian learning via stochastic gradient Langevin dynamics
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
Develops robust SGLD with non-asymptotic convergence bounds for non-convex DRO and applies it to neural network regression under adversarial corruption.
citing papers explorer
-
On quantitative Laplace-type convergence results for some exponential probability measures, with two applications
Establishes W1 quantitative bounds for Laplace-type convergence of measures with norm-like potentials using coarea formula under generalized Jacobian invertibility, applied to maxent and SGLD.
-
Robust SGLD algorithm for solving non-convex distributionally robust optimisation problems
Develops robust SGLD with non-asymptotic convergence bounds for non-convex DRO and applies it to neural network regression under adversarial corruption.