Bayesian computation: a perspective on the current state, and sampling backwards and forwards
arXiv:1502.01148
Abstract
The past decades have seen enormous improvements in computational inference based on statistical models, with continual enhancement in a wide range of computational tools, in competition. In Bayesian inference, first and foremost, MCMC techniques continue to evolve, moving from random walk proposals to Langevin drift, to Hamiltonian Monte Carlo, and so on, with both theoretical and algorithmic inputs opening wider access to practitioners. However, this impressive evolution in capacity is confronted by an even steeper increase in the complexity of the models and datasets to be addressed. The difficulties of modelling and then handling ever more complex datasets most likely call for a new type of tool for computational inference that dramatically reduce the dimension and size of the raw data while capturing its essential aspects. Approximate models and algorithms may thus be at the core of the next computational revolution.
30 pages (incl. 8 pages of references), 5 figures
References in corpus (34)
- An efficient, multiple range random walk algorithm to calculate the density of states
- Approximate Bayesian computation scheme for parameter inference and model selection in dynamical systems
- Expectation Propagation for approximate Bayesian inference
- General state space Markov chains and MCMC algorithms
- Inferring population history with DIYABC: a user-friendly approach to Approximate Bayesian Computation
- Non-linear regression models for Approximate Bayesian Computation
- A Comparative Review of Dimension Reduction Methods in Approximate Bayesian Computation
- Generalized Forward-Backward Splitting
- A General Framework for the Parametrization of Hierarchical Models
- Convex Optimization for Big Data
- On the ergodicity properties of some adaptive MCMC algorithms
- A Short History of Markov Chain Monte Carlo: Subjective Recollections from Incomplete Data
- MCMC for doubly-intractable distributions
- Approximate Bayesian computation (ABC) gives exact results under the assumption of model error
- Computational and Statistical Tradeoffs via Convex Relaxation
- Equi-energy sampler with applications in statistical inference and statistical mechanics
- On the efficiency of pseudo-marginal random walk Metropolis algorithms
- Speeding Up MCMC by Efficient Data Subsampling
- Optimal scalings for local Metropolis--Hastings chains on nonproduct targets in high dimensions
- On Russian Roulette Estimates for Bayesian Inference with Doubly-Intractable Likelihoods
- Robust and Scalable Bayes via a Median of Subset Posterior Measures
- Parallelizing MCMC via Weierstrass Sampler
- On the ergodicity of the adaptive Metropolis algorithm on unbounded domains
- Multiprocess parallel antithetic coupling for backward and forward Markov Chain Monte Carlo
- Optimizing The Integrator Step Size for Hamiltonian Monte Carlo
- Variance bounding and geometric ergodicity of Markov chain Monte Carlo kernels for approximate Bayesian computation
- Ergodicity of Approximate MCMC Chains with Applications to Large Data Sets
- Optimal scaling for the transient phase of Metropolis Hastings algorithms: The longtime behavior
- Bayesian Computation and Model Selection in Population Genetics
- Consistency and fluctuations for stochastic gradient Langevin dynamics
- Unbiased Bayes for Big Data: Paths of Partial Posteriors
- Individual adaptation: an adaptive MCMC scheme for variable selection problems
- Efficient implementation of Markov chain Monte Carlo when using an unbiased likelihood estimator
- Stability of Noisy Metropolis-Hastings