diff --git a/scenarios/domains/crossing/p2/rexd_config.cfg b/scenarios/domains/crossing/p2/rexd_config.cfg index 76f1d9176b366e92b17515b0da61f788f0d8a87d..e78e5f67f7f7437b182aebb085ced00eb55f59e4 100644 --- a/scenarios/domains/crossing/p2/rexd_config.cfg +++ b/scenarios/domains/crossing/p2/rexd_config.cfg @@ -21,7 +21,7 @@ exploration_known_threshold = 2 confirm_dangerous_actions = false -use_rmax = true +use_rmax = false use_vmin = false rmax_reward = 15 vmin_value = 0.1 diff --git a/scenarios/domains/crossing/p3/rexd_config.cfg b/scenarios/domains/crossing/p3/rexd_config.cfg index 111c02ef6f3203f5d1a8dc493ce369705d93e751..6f7b9234e39e0398f6f4b3feadfe073dded94d4d 100644 --- a/scenarios/domains/crossing/p3/rexd_config.cfg +++ b/scenarios/domains/crossing/p3/rexd_config.cfg @@ -21,7 +21,7 @@ exploration_known_threshold = 2 confirm_dangerous_actions = false -use_rmax = true +use_rmax = false use_vmin = false rmax_reward = 15 vmin_value = 0.1 @@ -34,8 +34,8 @@ learner = lfit learner_path = lfit lfit_learner_max_action_variables = 3 lfit_learner_conflicts_avoid_all = false -lfit_learner_conflicts_heuristic_max_iterations = 400 -lfit_learner_conflicts_heuristic_max_rules_per_iter = 1000 +lfit_learner_conflicts_heuristic_max_iterations = 200 +lfit_learner_conflicts_heuristic_max_rules_per_iter = 600 lfit_learner_score_optimistic_value = 0.95 lfit_learner_score_use_confidence = true lfit_learner_score_likelihood_scaling = 1.0