Group Norm for Learning Structured SVMs with Unstructured Latent Variables
Name
Freeman_Group norm.pdf
Size
7.12 MB
Format
Adobe PDF
Checksum (MD5)
98110d248ede2038ffe9c625224868b8
Author(s)
Chen, Daozheng
Batra, Dhruv
Freeman, William T.
Date Issued
December 2013
Journal
Proceedings of the 2013 IEEE International Conference on Computer Vision
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Citation
Chen, Daozheng, Dhruv Batra, and William T. Freeman. “Group Norm for Learning Structured SVMs with Unstructured Latent Variables.” 2013 IEEE International Conference on Computer Vision (December 2013).
Version
Author's final manuscript
Abstract
Latent variables models have been applied to a number of computer vision problems. However, the complexity of the latent space is typically left as a free design choice. A larger latent space results in a more expressive model, but such models are prone to over fitting and are slower to perform inference with. The goal of this paper is to regularize the complexity of the latent space and learn which hidden states are really relevant for prediction. Specifically, we propose using group-sparsity-inducing regularizers such as ℓ[subscript 1]-ℓ[subscript 2] to estimate the parameters of Structured SVMs with unstructured latent variables. Our experiments on digit recognition and object detection show that our approach is indeed able to control the complexity of latent space without any significant loss in accuracy of the learnt model.
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1109/ICCV.2013.58
