Q-Learning Scheduler for Multi Task Learning Through the use of Histogram of Task Uncertainty
Kourosh Meshgi, Maryam Sadat Mirzaei, Satoshi Sekine · 2022
Simultaneous training of a multi-task learning (MTL) network on different domains or tasks is not always straightforward.It could lead to inferior performance or generalization compared to the corresponding single-task networks.An effective training scheduling method is deemed necessary to maximize the benefits of multitask learning.Traditional schedulers follow a heuristic or prefixed strategy, ignoring the relation of the tasks, their sample complexities, and the state of the emergent shared features.We proposed a deep Q-Learning Scheduler (QLS) that monitors the state of the tasks and the shared features using a novel histogram of task uncertainty, and through trial-and-error, learns an optimal policy for task scheduling.Extensive experiments on multi-domain and multitask settings with various task difficulty profiles have been conducted, the proposed method is benchmarked against other schedulers, its superior performance has been demonstrated, and results are discussed.