arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1608.06374cs.LGcs.CV
收藏

深度双稀疏编码器:不仅学习稀疏化特征,还学习稀疏化参数

Deep Double Sparsity Encoder: Learning to Sparsify Not Only Features But Also Parameters

University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)

Zhangyang Wang, Thomas S. Huang

摘要:

本文强调了在深度建模的背景下,联合开发问题结构和参数结构的重要性。作为一个具体而有趣的例子,我们描述了深度双稀疏编码器(DDSE),它受到字典学习的双稀疏模型的启发。DDSE同时稀疏的输出功能和学习的模型参数,在一个统一的框架。除了直观的模型解释,DDSE还具有紧凑的模型大小和低复杂度。大量的模拟比较DDSE与几个精心设计的基线,并验证DDSE一贯优越的性能。我们进一步将DDSE应用于大脑编码的新应用领域,取得了可喜的初步成果。

英文摘要:

This paper emphasizes the significance to jointly exploit the problem structure and the parameter structure, in the context of deep modeling. As a specific and interesting example, we describe the deep double sparsity encoder (DDSE), which is inspired by the double sparsity model for dictionary learning. DDSE simultaneously sparsities the output features and the learned model parameters, under one unified framework. In addition to its intuitive model interpretation, DDSE also possesses compact model size and low complexity. Extensive simulations compare DDSE with several carefully-designed baselines, and verify the consistently superior performance of DDSE. We further apply DDSE to the novel application domain of brain encoding, with promising preliminary results achieved.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑