Sampler roadmap / overview

[MichaelClerx]

Can't find an existing ticket for this, though sure we had one at some point.

Classed as black, blue, and red in Ben's diagram

Likelihood-free

• [ ] Hamiltonian ABC
• [ ] Markov chain monte-carlo ABC --> TICKET OR PR NEEDED
• [ ] Regression-adjustment ABC
• [x] Rejection ABC --> PR ~#881 #925~ #1442
• [x] Sequential monte-carlo ABC PR ~#1055~ #1442

Derivative-free

• [x] AM global-adaptation MCMC --> PR #491
• [x] AM local-componentwise MCMC --> PR #491
• [x] Delayed rejection adaptive metropolis (DRAM) --> PR #491
• [x] Differential evolution --> DifferentialEvolutionMCMC
• [x] Differential evolution adaptive metropolis (DREAM) --> DreamMCMC
• [x] Ellipsoidal nested sampling --> NestedEllipsoidSampler
• [x] Emcee hammer --> EmceeHammerMCMC
• [ ] Generalised elliptical slice sampling --> PR #900
• [ ] Gradient-free HMC #101
• [ ] Localised adaptive covariance --> PR #491
• [ ] Mode jumping adaptive MCMC #492
• [x] Population MCMC --> PopulationMCMC
• [x] Rao-blackwell AM MCMC --> PR #491
• [x] Random walk metropolis --> MetropolisRandomWalkMCMC
• [x] Remi AM MCMC --> AdaptiveCovarianceMCMC
• [x] Rejection nested sampling --> NestedRejectionSampler
• [ ] Sequential Monte Carlo --> PR #749
• [x] Slice sampling with doubling --> SliceDoublingMCMC
• [ ] Slice sampling with hyperrectangles --> PR #895
• [ ] Slice sampling with overrelaxation --> Temp PR #863
• [ ] Slice sampling with Peano curves
• [x] Slice sampling with stepout --> SliceStepoutMCMC
• [ ] Transport map accelerated MCMC
• [x] Vanilla AM MCMC --> PR #491
• [ ] Vina copula adaptive MCMC Ticket #495

1st order sensitivities

• [ ] Adaptive HMC
• [ ] Adaptive MALA #adapti
• [ ] Bouncy particle HMC
• [ ] Covariance matching SS --> PR #896
• [ ] Diffusive nested sampling #152
• [ ] Dynamic nested sampling
• [ ] dyPolychord
• [ ] Galilean monte carlo #341
• [x] Hamiltonian Monte Carlo (HMC) --> HamiltonianMCMC
• [x] Metropolis-adjusted langevin algorithm (MALA) --> MALAMCMC
• [x] Monomial gamma HMC --> MonomialGammaHamiltonianMCMC
• [ ] Multinest #1034
• [ ] No-U-Turn sampler (NUTS) #1031
• [ ] Polychord #305
• [x] Rank shrinking SS --> #899
• [ ] Shadow HMC #521

2nd order sensitivities

• [ ] Riemannian manifold HMC
• [ ] Riemannian manifold MALA

Framework for all sampling (temper distribution, sample from it, reweight via importance sampling):

• [ ] SIS MCMC --> PR #760

[MichaelClerx]

@ben could you check if the bottom 6 are e.g. different names for ones above? Found them on github but not in the pyramid

[MichaelClerx]

Aha, here it is: https://github.com/orgs/pints-team/projects/1#card-15881685

[ben18785]

Hi Michael,

Thanks! I've edited the comment to remove those methods I think are duplicates. Not sure what to call SIS MCMC -- it's an approach that can be applied to all samplers (and likelihood-based optimisers) essentially. Just involves heating the distribution, sampling from it, then reweighting based on importance weights. Can leave it as it stands for now though!

[MichaelClerx]

Thanks!