arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1905.12874cs.LGstat.ML
收藏

鲁棒深度表示的信息论学习

Information theoretic learning of robust deep representations

University Pierre et Marie Curie(皮埃尔与玛丽·居里大学);UPMC(巴黎第六大学(皮埃尔与玛丽·居里大学))

Nicolas Pinchaud

摘要:

我们提出了一种新的目标函数,用于学习基于信息论的数据的鲁棒深度表示。将数据投影到特征向量空间中,使得相对于监控信号的所有特征子集的互信息最大化。该目标函数通过保存相对于在面对噪声或不可用的特征时的监督的可用信息来产生鲁棒的表示。虽然目标函数不是直接易处理的,但我们能够推导出替代目标函数。最小化这种替代损失鼓励特征相对于监督信号是非冗余的并且有条件地独立。为了评估所获得的解决方案的质量,我们已经进行了一组初步的实验,显示出有希望的结果。

英文摘要:

We propose a novel objective function for learning robust deep representations of data based on information theory. Data is projected into a feature-vector space such that the mutual information of all subsets of features relative to the supervising signal is maximized. This objective function gives rise to robust representations by conserving available information relative to supervision in the face of noisy or unavailable features. Although the objective function is not directly tractable, we are able to derive a surrogate objective function. Minimizing this surrogate loss encourages features to be non-redundant and conditionally independent relative to the supervising signal. To evaluate the quality of obtained solutions, we have performed a set of preliminary experiments that show promising results.

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。

↑