bandit_epsilon
Documentation
{"subtype": "greedy", "historical_info": {"arms_sampled": {"arm2": {"win": 20, "total": 30, "loss": 0}, "arm3": {"win": 0, "total": 0, "loss": 0}, "arm1": {"win": 20, "total": 25, "loss": 0}}}, "hyperparameter_info": {"epsilon": 0.05}}
Submit