How these researchers tried something unconventional to come out with a smaller yet better Image…
Mate Labs

Thanks for the good write-up. I’m not sure if this statement is correct:

> Also it can reduce the number of parameters and computations in the network, therefore, controlling things like overfitting.

Doesn’t replacing the pooling layers with conv layers add more parameters? The pooling layers have no parameters whereas the convolutional layers do.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.