5.4. COMPETITION ENTRIES 91
5.4.1 RANDOM AGENT (SAMPLE AGENTS)
5.4.2 DRL ALGORITHMS (SAMPLE AGENTS)
5.4.3 MULTI-ARMED BANDIT ALGORITHM
DontUnderestimateUchiha
T best 1
T
best T
0
0:5
T
D
0
0:0001T
8T 2 f1; : : : ; 2000g 0:5
T
0:3
40%
5.4.4 SARSA
sampleLearner ercumentilhan fraBot-RL-Sarsa
sampleLearner ercumentil-
han
fraBot-RL-Sarsa
48