Nan Ma

TL;DR: Li et al. as discussed by the authors proposed a label correlation network to explore the label relation knowledge with flexible aggregators, and then they completed the label on unlabeled instances in a semi-supervised fashion.

TL;DR: It is demonstrated that distinguishing request scenarios helps provide preferable context-aware route recommendations, and an approach via a combination of the weighted shortest path problem and deep learning is designed and implemented, and evaluated using real-world transportation data.

TL;DR: An instance reweighting strategy is employed to cope with severe label noise in comment data, where the weights for harmful noisy instances are small and strong criteria like AUC rather than accuracy and the validation performance are optimized for the correction of biased data label.

TL;DR: In this paper, the authors focus on the complete dispatch for the ride-hailing platform and propose a network flow accelerated algorithm to obtain the dispatch policy when perfect information is available.