Beyond Shared Hierarchies: Deep Multitask Learning through Soft Layer Ordering
This work addresses a bottleneck in deep multitask learning for AI researchers, offering an incremental improvement over existing methods.
The paper tackled the limitation of parallel ordering in deep multitask learning by proposing a soft ordering approach, which outperformed existing methods across multiple domains by enabling more effective sharing of layers.
Existing deep multitask learning (MTL) approaches align layers shared between tasks in a parallel ordering. Such an organization significantly constricts the types of shared structure that can be learned. The necessity of parallel ordering for deep MTL is first tested by comparing it with permuted ordering of shared layers. The results indicate that a flexible ordering can enable more effective sharing, thus motivating the development of a soft ordering approach, which learns how shared layers are applied in different ways for different tasks. Deep MTL with soft ordering outperforms parallel ordering methods across a series of domains. These results suggest that the power of deep MTL comes from learning highly general building blocks that can be assembled to meet the demands of each task.