arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1503.03438cs.LGcs.NEstat.ML
收藏

复值卷积网络的数学动机

A mathematical motivation for complex-valued convolutional networks

Facebook Artificial Intelligence Research(脸书人工智能研究院)

Joan Bruna, Soumith Chintala, Yann LeCun, Serkan Piantino, Arthur Szlam, Mark Tygert

摘要:

复值卷积网络(convnet)实现了以下三个操作的组合的重复应用,递归地将组合应用于非负实数的输入向量:(1)与复值向量的卷积,然后是(2)取结果向量的每个条目的绝对值,然后是(3)局部平均。对于处理实值随机向量,复值卷积网可以被视为“数据驱动的多尺度加窗功率谱”、“数据驱动的多尺度加窗绝对谱”、“数据驱动的多小波绝对值”或(在其最一般的配置中)“数据驱动的非线性多小波包”。“事实上,当convnet滤波器是窗口化复值指数时,复值convnet可以计算多尺度窗口化频谱。标准实值convnets,使用整流线性单元(ReLU),sigmoidal(例如,logistic或tanh)非线性,最大。汇集等,并没有明显地表现出与数据驱动小波相同的精确对应(而对于复值convnets,对应关系不仅仅是一个模糊的类比)。由于精确对应,小波的数学分析非常丰富和严格,直接适用于(复值)convnets。

英文摘要:

A complex-valued convolutional network (convnet) implements the repeated application of the following composition of three operations, recursively applying the composition to an input vector of nonnegative real numbers: (1) convolution with complex-valued vectors followed by (2) taking the absolute value of every entry of the resulting vectors followed by (3) local averaging. For processing real-valued random vectors, complex-valued convnets can be viewed as "data-driven multiscale windowed power spectra," "data-driven multiscale windowed absolute spectra," "data-driven multiwavelet absolute values," or (in their most general configuration) "data-driven nonlinear multiwavelet packets." Indeed, complex-valued convnets can calculate multiscale windowed spectra when the convnet filters are windowed complex-valued exponentials. Standard real-valued convnets, using rectified linear units (ReLUs), sigmoidal (for example, logistic or tanh) nonlinearities, max. pooling, etc., do not obviously exhibit the same exact correspondence with data-driven wavelets (whereas for complex-valued convnets, the correspondence is much more than just a vague analogy). Courtesy of the exact correspondence, the remarkably rich and rigorous body of mathematical analysis for wavelets applies directly to (complex-valued) convnets.

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。

↑