Key Problems

• CNN implicitly makes the assumption that the latent concepts are linearly separable
• the data for the same concept often live on a nonlinear manifold, therefore the representations that capture these concepts are generally highly nonlinear function of the input

Contributions

• enhance model discriminability for local patches within the receptive field
• utilize global average pooling over feature maps in the classification layer, which is easier to interpret and less prone to overfitting

Methods

Linear convolution layer

MLP Convolution Layers

Architecture

Global Average Pooling

• This structure bridges the convolutional structure with traditional neural network classifiers.
• It treats the convolutional layers as feature extractors, and the resulting feature is classified in a traditional way.
• t has improved the generalization ability and largely prevents overfitting

Experiments