IoT-Based Reinforcement Learning Using Probabilistic Model for Determining Extensive Exploration through Computational Intelligence for Next-Generation Techniques

Table 3

Cumulative regret of each algorithm in 10,000 episodes.

Variable settings		DQN	BDQN	BBDQN
K = 10	N = 10	9914.88	599.65	649.29
	N = 20	9922.02	9914.60	1590.03
	N = 30	9912.36	9910.96	5203.65
N = 20	K = 10	9922.02	9914.02	1590.03
	K = 20	9922.02	9912.13	1901.18
	K = 30	9922.02	1857.49	1784.71