Figure 5

Demonstration of order recognition in 16 slot machine environments. (a) (i) Reward probability arrangement and (ii) its sorted representation. (b) The accumulated incidence pattern of the ranking-and-value matrix after the loop of (i) 1, (ii) 50, (iii) 100, (iv) 200, (v) 500, and (vi) 1000. (c) The time evolution of the estimated ranking of each slot machine. (d) The number of correctly ordered machines over time.