Learning A Deep ℓ_∞ Encoder for Hashing
release_isspjfxi6bf3lkrknx6nfrxzqq
by
Zhangyang Wang, Yingzhen Yang, Shiyu Chang, Qing Ling, Thomas S. Huang
2016
Abstract
We investigate the ℓ_∞-constrained representation which
demonstrates robustness to quantization errors, utilizing the tool of deep
learning. Based on the Alternating Direction Method of Multipliers (ADMM), we
formulate the original convex minimization problem as a feed-forward neural
network, named Deep ℓ_∞ Encoder, by introducing the novel
Bounded Linear Unit (BLU) neuron and modeling the Lagrange multipliers as
network biases. Such a structural prior acts as an effective network
regularization, and facilitates the model initialization. We then investigate
the effective use of the proposed model in the application of hashing, by
coupling the proposed encoders under a supervised pairwise loss, to develop a
Deep Siamese ℓ_∞ Network, which can be optimized from end to
end. Extensive experiments demonstrate the impressive performances of the
proposed model. We also provide an in-depth analysis of its behaviors against
the competitors.
In text/plain
format
Archived Files and Locations
application/pdf 683.1 kB
file_7mqd7m3i5bhslmks2io47vymoa
|
arxiv.org (repository) web.archive.org (webarchive) |
1604.01475v1
access all versions, variants, and formats of this works (eg, pre-prints)