- initial commit

Author: MathPhysSim

README.md

@@ -8,8 +8,10 @@ A new implementation of the NAF with doule Q learning:
 A new implementation of a AE-dyna:
 ![AE-DYNA](Figures/AE-DYNA_observables.png)
 ![AE-DYNA](Figures/AE-DYNA_verification.png)
+
 A variant of the ME-TRPO:
 ![ME-TRPO](Figures/ME-TRPO_observables.png)
 ![ME-TRPO](Figures/ME-TRPO_verification.png)
+
 The evolution:
 ![ME-TRPO](Figures/Learning_evolution.png)