Time-Driven Scheduling Based on Reinforcement Learning for Reasoning Tasks in Vehicle Edge Computing

Table 1

Notations and descriptions.

Notation	Description
, ,	A reasoning task, a set of subtasks, and the data dependencies between subtasks
	Computational scheduling plan for tasks
, ,	Task size, computation intensity, tolerable delay of subtask
,	Computation latency in vehicle and RSU
,	Local computing capacity and scheduling capacity
	Transmission latency of subtasks
	Completion latency of a reasoning task
, ,	State, action, and reward of an MDP at time step t
,	Parametrized policy and value function for computation scheduling