Is the submission that taking advantage of data leakage be considered as valid result? Say, after many random tryings we get a very good results?
Posted by: HEYUXUAN @ Jan. 10, 2023, 8:46 a.m.I think actions should be computed by a reinforcement learning agent, rather than other methods.
Posted by: yj @ Jan. 10, 2023, 9:37 p.m.