rpl_e function

Step 4: Replaying the experiment with optimal parameters