作者提出的Context-Gated Convolution,把卷积层当做一个“自适应的处理器”,可以根据图像中的语义信息来调整卷积核的权重。 这个方法实现起来并不容易,因为对于输入feature map 的尺寸为(c,h,w)(c,h,w), 输出 feature map 的尺寸为(o,h,w)(o,h,w),这样,卷积参数量就是o×c×k×ko×c×k×k。所以,...
这些方法仍没有办法对卷积核建模做到“changing the structure of correlations over neuronal ensembles”。 作者提出的Context-Gated Convolution,把卷积层当做一个“自适应的处理器”,可以根据图像中的语义信息来调整卷积核的权重。 这个方法实现起来并不容易,因为对于输入feature map 的尺寸为(c,h,w), 输出 feature...
Cross-modal context-gated convolution also brings more possibilities to the layer design for multi-modal sequential modeling. Experiments on multi-modal sentiment analysis datasets under both word-aligned and unaligned conditions show the validity of our approach....
Motivated by this, we propose one novel Context-Gated Convolution (CGC) to explicitly modify the weights of convolutional layers adaptively under the guidance of global context. As such, being aware of the global context, the modulated convolution kernel of our proposed CGC can better extract ...