| model,average_reward,win_rate,loss_rate,draw_rate,eval_episodes | |
| best_basic_count,-0.008698,0.433611,0.481673,0.084716,1000000 | |
| five_m_expert_prior,-0.008698,0.433611,0.481673,0.084716,1000000 | |
| five_m_aggressive,-0.024801,0.430806,0.482306,0.086888,1000000 | |