AdaShare: Learning What To Share For Efficient Deep Multi-Task Learning
This work addresses the challenge of efficient multi-task learning for computer vision applications, offering a novel method to dynamically share network layers across tasks.
The paper tackles the problem of multi-task learning in computer vision by proposing AdaShare, an adaptive sharing approach that learns which layers to share across tasks to optimize recognition accuracy and resource efficiency, achieving improved performance over state-of-the-art methods on diverse benchmark datasets.
Multi-task learning is an open and challenging problem in computer vision. The typical way of conducting multi-task learning with deep neural networks is either through handcrafted schemes that share all initial layers and branch out at an adhoc point, or through separate task-specific networks with an additional feature sharing/fusion mechanism. Unlike existing methods, we propose an adaptive sharing approach, called AdaShare, that decides what to share across which tasks to achieve the best recognition accuracy, while taking resource efficiency into account. Specifically, our main idea is to learn the sharing pattern through a task-specific policy that selectively chooses which layers to execute for a given task in the multi-task network. We efficiently optimize the task-specific policy jointly with the network weights, using standard back-propagation. Experiments on several challenging and diverse benchmark datasets with a variable number of tasks well demonstrate the efficacy of our approach over state-of-the-art methods. Project page: https://cs-people.bu.edu/sunxm/AdaShare/project.html.