set experimental config

config.lua

@@ -50,19 +50,27 @@ local common_cfg = {
 }
 
 local cfg = {
-    -- best performance so far, albeit a bit slow to train.
+    log_fn = 'logs-xnes.csv',
-    -- NOTE: uncomment for original values used.
+    params_fn = 'params-xnes.txt',
-    log_fn = 'log.csv', -- can be nil to disable logging.
+
-    params_fn = nil, -- can be nil to generate based on param count.
+    decrement_reward = true,
+    score_multiplier = 5,
+
+    starting_world = 1,
+    starting_level = 1,
     starting_lives = 1,
+    cap_time = 300,
+
     deterministic = true,
-    layernorm = true,
+
-    epoch_trials = 24, --20,
+    epoch_trials = 50,
-    epoch_top_trials = 18, --10,
+    epoch_top_trials = 9999,
-    ars_lips = true,
+    negate_trials = false,
-    deviation = 0.05, --0.1,
+
-    weight_decay = 0.0002, --0.0004,
+    es = 'xnes',
-    learning_rate = 2.0, --1.0
+    learning_rate = 0.14,
+    deviation = 1.0,
+    weight_decay = 0.0,
 }
 
 -- TODO: so, uhh..