A Unified Perspective on Multi-Domain and Multi-Task Learning
This work addresses the challenge of adapting models to new tasks or domains without training data, which is incremental as it builds on existing MTL/MDL methods.
The paper tackles the problem of unifying multi-task and multi-domain learning by introducing a semantic descriptor framework, which also enables zero-shot learning and zero-shot domain adaptation, with experiments showing it outperforms various alternatives.
In this paper, we provide a new neural-network based perspective on multi-task learning (MTL) and multi-domain learning (MDL). By introducing the concept of a semantic descriptor, this framework unifies MDL and MTL as well as encompassing various classic and recent MTL/MDL algorithms by interpreting them as different ways of constructing semantic descriptors. Our interpretation provides an alternative pipeline for zero-shot learning (ZSL), where a model for a novel class can be constructed without training data. Moreover, it leads to a new and practically relevant problem setting of zero-shot domain adaptation (ZSDA), which is the analogous to ZSL but for novel domains: A model for an unseen domain can be generated by its semantic descriptor. Experiments across this range of problems demonstrate that our framework outperforms a variety of alternatives.