Thank you for your detailed reviews and comments. Wehope our clarifications, which we will include in the final1 version of the paper, will strengthen your confidence in the novelty and significance of our results. Both speed up DPP sampling given a polynomial time pre-processing step.

Erd os GNN [29] is a novel framework with unsupervised learning, however, its main limitation is that this framework is incapable of handling constraints beyond simple node constraints. Following the implementation from [39], the job nodes are scheduled in sequential order with PPO-Single. Herestateisthe current DAGGk with atimestamp, and some of the nodes are already scheduled by the current timestamp. Torepresent the current state of the problem, the finished nodes, running nodes and unscheduled nodes are marked by different node 15 attributes, so that the state information is fully encoded by the nodes and edges ofGk. After anode finishes, itwillfreesome resources, and sometimes add some available nodes to be scheduled.

Example Reweightingdirectly assigns an importance weight to the standard CE training loss, accordingtothebiasdegreeβ: Lreweight = (1 β)y logpm (3) Confidence Regularizationis based on knowledge distillation [9]. It involves a teacher model trainedwiththestandardCEloss. Specifically, we calculate the weighted average of the F1 score of each class. The splits used for evaluation are highlightedwithredcolor. To address this problem, we select the best checkpoint after0.7 tmax of training, butstill according to the performance on the ID devset.