Class markov::HillClimbingWalker

A Hill climbing walker accepts every move reducing the cost and keeps moving if cost is same. NOTE: computing cost_difference is typically not superfluous as it will be needed within Walker::make_step() to update the cached cost_.

Inheritance

markov::Walker

markov::HillClimbingWalker

Inherited Members

get_lowest_state

check_lowest

state

reset_evaluation_counter

make_sweep

swap_state

make_sweep

set_model

model

Walker

make_sweeps

make_sweep

attempt_transition

get_evaluation_counter

set_rng

save_lowest

rng

memory_estimate

apply_transition

init

get_lowest_cost

make_step

cost

init

apply_scale_factor

compare

attempt_transition

apply_transition

verify_cost_difference

configure

render

~Component

Component

get_status

param

get_class_name

Methods

accept()

Implementations of the walker interface must provide a method to decide which steps to accept/reject. Currently this is only based on cost_difference, but the state or transition could be added.

Declaration

bool markov::HillClimbingWalker<Model>::accept(const typename Model::Cost_T&cost_diff) override