Cross-Stitch Networks for Multi-Task Learning

Ishan Misra, Abhinav Shrivastava, Abhinav Gupta, Martial Hebert; Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 3994-4003


Multi-task learning in Convolutional Networks has displayed remarkable success in the field of recognition. This success can be largely attributed to learning shared representations from multiple supervisory tasks. However, existing multi-task approaches rely on enumerating multiple network architectures specific to the tasks at hand, that do not generalize. In this paper, we propose a principled approach to learn shared representations in ConvNets using multi-task learning. Specifically, we propose a new sharing unit: "cross-stitch" unit. These units combine the activations from multiple networks and can be trained end-to-end. A network with cross-stitch units can learn an optimal combination of shared and task-specific representations. Our proposed method generalizes across multiple tasks and shows dramatically improved performance over baseline methods for categories with few training examples.

Related Material

[pdf] [video]
author = {Misra, Ishan and Shrivastava, Abhinav and Gupta, Abhinav and Hebert, Martial},
title = {Cross-Stitch Networks for Multi-Task Learning},
booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2016}