arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1604.06153cs.LG
收藏

非广延信息理论机

Nonextensive information theoretical machine

Graduate School at Shenzhen, Tsinghua University(清华大学深圳研究生院)

Chaobing Song, Shu-Tao Xia

摘要:

本文基于Shannon信息论的非广延推广,提出了一种新的判别模型\emph{nonextensive information theoretical machine (NITM)}。在NITM中,权重参数被视为随机变量。采用Tsallis散度对权值参数的分布进行正则化,采用最大非归一化Tsallis熵分布对拟合效果进行评价。一方面,证明了一些著名的基于边际的损失函数,如$\ell_{0/1}$损失、铰链损失、平方铰链损失和指数损失等可以用非归一化Tsallis熵统一起来。另一方面,将高斯先验正则化推广到具有相似计算复杂性的Student-t先验正则化。该模型可以有效地解决基于梯度的凸优化,其性能在标准数据集上进行了说明。

英文摘要:

In this paper, we propose a new discriminative model named \emph{nonextensive information theoretical machine (NITM)} based on nonextensive generalization of Shannon information theory. In NITM, weight parameters are treated as random variables. Tsallis divergence is used to regularize the distribution of weight parameters and maximum unnormalized Tsallis entropy distribution is used to evaluate fitting effect. On the one hand, it is showed that some well-known margin-based loss functions such as $\ell_{0/1}$ loss, hinge loss, squared hinge loss and exponential loss can be unified by unnormalized Tsallis entropy. On the other hand, Gaussian prior regularization is generalized to Student-t prior regularization with similar computational complexity. The model can be solved efficiently by gradient-based convex optimization and its performance is illustrated on standard datasets.

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。

↑