A Unified Perspective on Multi-Domain and Multi-Task Learning

@article{Yang2014AUP,
  title={A Unified Perspective on Multi-Domain and Multi-Task Learning},
  author={Yongxin Yang and Timothy M. Hospedales},
  journal={CoRR},
  year={2014},
  volume={abs/1412.7489}
}
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… CONTINUE READING
Highly Cited
This paper has 75 citations. REVIEW CITATIONS
Related Discussions
This paper has been referenced on Twitter 14 times. VIEW TWEETS

Citations

Publications citing this paper.
Showing 1-10 of 50 extracted citations

Multi-Task Learning with Low Rank Attribute Embedding for Multi-Camera Person Re-Identification

Chi Su, Fan Yang, +3 authors Wen Gao
IEEE Transactions on Pattern Analysis and Machine Intelligence • 2018
View 4 Excerpts
Highly Influenced

Learning a Deep Embedding Model for Zero-Shot Learning

2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) • 2017
View 8 Excerpts
Highly Influenced

75 Citations

010203020152016201720182019
Citations per Year
Semantic Scholar estimates that this publication has 75 citations based on the available data.

See our FAQ for additional information.

References

Publications referenced by this paper.
Showing 1-10 of 35 references

Learning to detect unseen object classes by between-class attribute transfer

2009 IEEE Conference on Computer Vision and Pattern Recognition • 2009
View 18 Excerpts
Highly Influenced

Convex multi-task feature learning

View 8 Excerpts
Highly Influenced

Zero-Shot Learning Through Cross-Modal Transfer

NIPS • 2013
View 4 Excerpts
Highly Influenced

Exploiting web images for event recognition in consumer videos: A multiple source domain adaptation approach

2012 IEEE Conference on Computer Vision and Pattern Recognition • 2012
View 4 Excerpts
Highly Influenced

Geodesic flow kernel for unsupervised domain adaptation

2012 IEEE Conference on Computer Vision and Pattern Recognition • 2012
View 4 Excerpts
Highly Influenced

Zero-data Learning of New Tasks

View 4 Excerpts
Highly Influenced

Regularized multi--task learning

View 5 Excerpts
Highly Influenced

Similar Papers

Loading similar papers…