A Flexible Reinforced Bin Packing Framework with Automatic Slack Selection

Table 2

Parameters and corresponding descriptions.

Parameter	Description
	Action taken at time step
	State at time step
	Reward at time step
	a set of actions
	a set of states
	Instant reward
	Transition probability between states
	The sum of reward at time step
	Learning rate
	The discount factor
	State-action value function
	Slack
	Number of bins used by the concrete algorithm
	Number of bins contained in the optimal solution of each bin packing instance
	Number of feasible optimal solution instances that the algorithm does not achieve
	Deviation percentage between the solution reached by each algorithm and the optimal solution
	Competition ratio