The MAP learner chooses uniformly among the languages that maximize prior times likelihood.
- mapLearner κ μ = ProbabilityTheory.Kernel.ofFunOfCountable fun (d : D) => ProbabilityTheory.uniformOn ↑(Finset.univ.argmax fun (h : H) => μ {h} * (κ h) {d})
Used by mapLearner_apply, instIsMarkovKernelMapLearnerOfMeasurableSingletonClassOfNonempty, mapLearner_prior_eq, mapLearner_comp_real_one_two, mapLearner_comp_real_two_one, invariant_mapLearner_comp_produce_iff, invariant_mapLearner_comp_produce_prior_iff
Used in BrochhagenFrankeVanRooij2018, KirbyDowmanGriffiths2007