Non-asymptotic error estimates for the Laplace approximation in Bayesian inverse problems.
Tapio HelinRemo KretschmannPublished in: Numerische Mathematik (2022)
Keyphrases
- error estimates
- inverse problems
- marginal likelihood
- error estimation
- cross validation
- model selection
- image reconstruction
- global optimization
- convex optimization
- expectation propagation
- hyperparameters
- optimization methods
- error analysis
- provably correct
- partial differential equations
- optimization problems
- early vision
- optimization algorithm
- smoothness constraint
- parameter space
- sample size
- genetic algorithm
- super resolution
- least squares
- feature vectors
- support vector
- computer vision