Learning Task Relatedness in Multi-Task Learning for Images in Context
This addresses the challenge of exploiting complex task relations in multi-task learning for multimedia applications, offering an incremental improvement over existing methods.
The paper tackles the problem of automatically learning task relatedness in multi-task learning when explicit domain knowledge is unavailable, introducing Selective Sharing to group tasks and share knowledge, resulting in consistent improvements in accuracy and parameter efficiency across five datasets.
Multimedia applications often require concurrent solutions to multiple tasks. These tasks hold clues to each-others solutions, however as these relations can be complex this remains a rarely utilized property. When task relations are explicitly defined based on domain knowledge multi-task learning (MTL) offers such concurrent solutions, while exploiting relatedness between multiple tasks performed over the same dataset. In most cases however, this relatedness is not explicitly defined and the domain expert knowledge that defines it is not available. To address this issue, we introduce Selective Sharing, a method that learns the inter-task relatedness from secondary latent features while the model trains. Using this insight, we can automatically group tasks and allow them to share knowledge in a mutually beneficial way. We support our method with experiments on 5 datasets in classification, regression, and ranking tasks and compare to strong baselines and state-of-the-art approaches showing a consistent improvement in terms of accuracy and parameter counts. In addition, we perform an activation region analysis showing how Selective Sharing affects the learned representation.