1. [Publications](/publications)
2. Auxiliary Learning by Implicit Differentiation
 
 # Auxiliary Learning by Implicit Differentiation

  ![](/sites/default/files/styles/wide/public/publications/framework.png?itok=p0gwuTW1)

 Training with multiple auxiliary tasks is a common practice used in deep learning for improving the performance on the main task of interest. Two main challenges arise in this multi-task learning setting: (i) Designing useful auxiliary tasks; and (ii) Combining auxiliary tasks into a single coherent loss. We propose a novel framework, AuxiLearn, that targets both challenges, based on implicit differentiation. First, when useful auxiliaries are known, we propose learning a network that combines all losses into a single coherent objective function. This network can learn non-linear interactions between auxiliary tasks. Second, when no useful auxiliary task is known, we describe how to learn a network that generates a meaningful, novel auxiliary task. We evaluate AuxiLearn in a series of tasks and domains, including image segmentation and learning with attributes. We find that AuxiLearn consistently improves accuracy compared with competing methods.



 ## Authors



Aviv Navon (BIU)

Idan Achituve (BIU)

[Haggai Maron](/person/haggai-maron)

[Gal Chechik](/person/gal-chechik)

Ethan Fetaya (BIU)

 

 

 ## Publication Date



Sunday, April 4, 2021

 

 ## Published in



[ICLR 2021](https://arxiv.org/abs/2007.02693)

 

 ## Research Area



[Artificial Intelligence and Machine Learning ](/research-area/machine-learning-artificial-intelligence)