bonus             E[reward] final   seeds with P(a0) > 0.5  median first sample of a0  final entropy  optimum P(a0)
none                        0.487                     0.27                      179.5          0.003          1.000
entropy 0.02                0.697                     0.57                      143.0          0.003          1.000
entropy 0.1                 0.997                     1.00                       27.5          0.031          0.997
entropy 0.5                 0.139                     0.00                       35.5          3.764          0.129
count 0.05                  0.510                     0.30                      179.5          0.003          1.000
count 0.2                   0.557                     0.37                      386.0          0.003          1.000
