Adaptive proposals

src/AdvancedMH.jl

@@ -9,7 +9,7 @@ import Random
 
 # Exports
 export MetropolisHastings, DensityModel, RWMH, StaticMH, StaticProposal, 
-    RandomWalkProposal, Ensemble, StretchProposal
+    RandomWalkProposal, Ensemble, StretchProposal, AdaptiveProposal
 
 # Reexports
 export sample, MCMCThreads, MCMCDistributed
@@ -104,5 +104,6 @@ end
 include("proposal.jl")
 include("mh-core.jl")
 include("emcee.jl")
+include("adaptive.jl")
 
 end # module AdvancedMH

src/adaptive.jl

@@ -0,0 +1,93 @@
+mutable struct Adaptor
+    accepted::Integer
+    total   ::Integer
+    tune    ::Integer   # tuning interval
+    target  ::Float64   # target acceptance rate
+    bound   ::Float64   # bound on logσ of Gaussian kernel
+    δmax    ::Float64   # maximum adaptation step
+end
+
+Adaptor(; tune=25, target=0.44, bound=10., δmax=0.2) =
+    Adaptor(0, 0, tune, target, bound, δmax)
+
+"""
+    AdaptiveProposal{P} 
+
+An adaptive Metropolis-Hastings proposal. In order for this to work, the
+proposal kernel should implement the `adapted(proposal, δ)` method, where `δ`
+is the increment/decrement applied to the scale of the proposal distribution
+during adaptation (e.g. for a Normal distribution the scale is `log(σ)`, so
+that after adaptation the proposal is `Normal(0, exp(log(σ) + δ))`).
+
+# Example
+```julia
+julia>  p = AdaptiveProposal(Uniform(-0.2, 0.2));
+
+julia> rand(p)
+0.07975590594518434
+```
+
+# References 
+
+Roberts, Gareth O., and Jeffrey S. Rosenthal. "Examples of adaptive MCMC."
+Journal of Computational and Graphical Statistics 18.2 (2009): 349-367.
+"""
+mutable struct AdaptiveProposal{P} <: Proposal{P}
+    proposal::P
+    adaptor ::Adaptor
+end
+
+function AdaptiveProposal(p; kwargs...) 
+    AdaptiveProposal(p, Adaptor(; kwargs...))
+end
+
+accepted!(p::AdaptiveProposal) = p.adaptor.accepted += 1
+accepted!(p::Vector{<:AdaptiveProposal}) = map(accepted!, p)
+accepted!(p::NamedTuple{names}) where names = map(x->accepted!(getfield(p, x)), names)
+
+# this is defined because the first draw has no transition yet (I think)
+propose(rng::Random.AbstractRNG, p::AdaptiveProposal, m::DensityModel) = 
+    rand(rng, p.proposal)
+
+# the actual proposal happens here
+function propose(
+    rng::Random.AbstractRNG,
+    proposal::AdaptiveProposal{<:Union{Distribution,Proposal}},
+    model::DensityModel,
+    t
+)
+    consider_adaptation!(proposal) 
+    t + rand(rng, proposal.proposal)
+end
+
+function q(proposal::AdaptiveProposal, t, t_cond) 
+    logpdf(proposal, t - t_cond)
+end
+
+function consider_adaptation!(p)
+    (p.adaptor.total % p.adaptor.tune == 0) && adapt!(p)
+    p.adaptor.total += 1
+end
+
+function adapt!(p::AdaptiveProposal)
+    a = p.adaptor
+    a.total == 0 && return 
+    δ  = min(a.δmax, 1. /√(a.total/a.tune))  # diminishing adaptation
+    α  = a.accepted / a.tune  # acceptance ratio
+    p_ = adapted(p.proposal, α > a.target ? δ : -δ, a.bound) 
+    a.accepted = 0 
+    p.proposal = p_
+end
+
+function adapted(d::Normal, δ, bound=Inf)
+    lσ = log(d.σ) + δ
+    lσ = abs(lσ) > bound ? sign(lσ) * bound : lσ
+    Normal(d.μ, exp(lσ))
+end
+
+function adapted(d::Uniform, δ, bound=Inf)
+    lσ = log(d.b) + δ
+    σ  = abs(lσ) > bound ? exp(sign(lσ) * bound) : exp(lσ)
+    Uniform(-σ, σ)
+end
+

src/mh-core.jl

@@ -208,8 +208,12 @@ function AbstractMCMC.step(
 
     # Decide whether to return the previous params or the new one.
     if -Random.randexp(rng) < logα
+        accepted!(spl.proposal)
         return params, params
     else
         return params_prev, params_prev
     end
 end
+
+function accepted!(p::P) where P<:Proposal end
+

src/proposal.jl

@@ -103,4 +103,4 @@ function q(
     t_cond
 )
     return q(proposal(t_cond), t, t_cond)
-end
+end

test/runtests.jl

@@ -11,7 +11,7 @@ using Test
     Random.seed!(1234)
 
     # Generate a set of data from the posterior we want to estimate.
-    data = rand(Normal(0, 1), 300)
+    data = rand(Normal(0., 1), 300)
 
     # Define the components of a basic model.
     insupport(θ) = θ[2] >= 0
@@ -52,6 +52,32 @@ using Test
         @test mean(chain2.μ) ≈ 0.0 atol=0.1
         @test mean(chain2.σ) ≈ 1.0 atol=0.1
     end
+
+    @testset "Adaptive random walk" begin
+        # Set up our sampler with initial parameters.
+        spl1 = MetropolisHastings([AdaptiveProposal(Normal(0,.4)), AdaptiveProposal(Normal(0,1.2))])
+        spl2 = MetropolisHastings((μ=AdaptiveProposal(Normal(0,1.4)), σ=AdaptiveProposal(Normal(0,0.2))))
+
+        # Sample from the posterior.
+        chain1 = sample(model, spl1, 100000; chain_type=StructArray, param_names=["μ", "σ"])
+        chain2 = sample(model, spl2, 100000; chain_type=StructArray, param_names=["μ", "σ"])
+
+        # chn_mean ≈ dist_mean atol=atol_v
+        @test mean(chain1.μ) ≈ 0.0 atol=0.1
+        @test mean(chain1.σ) ≈ 1.0 atol=0.1
+        @test mean(chain2.μ) ≈ 0.0 atol=0.1
+        @test mean(chain2.σ) ≈ 1.0 atol=0.1
+    end
+
+    @testset "Compare adaptive to simple random walk" begin
+        data = rand(Normal(2., 1.), 500)
+        m1 = DensityModel(x -> loglikelihood(Normal(x,1), data))
+        p1 = RandomWalkProposal(Normal())
+        p2 = AdaptiveProposal(Normal())
+        c1 = sample(m1, MetropolisHastings(p1), 10000; chain_type=Chains)
+        c2 = sample(m1, MetropolisHastings(p2), 10000; chain_type=Chains)         
+        @test ess(c2).nt.ess > ess(c1).nt.ess
+    end
 
     @testset "parallel sampling" begin
         spl1 = StaticMH([Normal(0,1), Normal(0, 1)])