hybrid of max pooling and average pooling

user_1866775 0
Viewed 2126

While tweaking a deep convolutional net using Keras (with the TensorFlow backend) I would like to try out a hybrid between MaxPooling2D and AveragePooling2D, because both strategies seem to improve two different aspects regarding my objective.

I'm thinking about something like this:

    -------
    |8 | 1|
x = ---+---
    |1 | 6|
    -------

average_pooling(x)                ->   4
max_pooling(x)                    ->   8
hybrid_pooling(x, alpha_max=0.0)  ->   4
hybrid_pooling(x, alpha_max=0.25) ->   5
hybrid_pooling(x, alpha_max=0.5)  ->   6
hybrid_pooling(x, alpha_max=0.75) ->   7
hybrid_pooling(x, alpha_max=1.0)  ->   8

Or as an equation:

hybrid_pooling(x, alpha_max) =
    alpha_max * max_pooling(x) + (1 - alpha_max) * average_pooling(x)

Since it looks like such a thing is not provided off the shelf, how can it be implemented in an efficient way?

spatial-pooling max-pooling keras tensorflow deep-learning
2 Answers

here an easy implementation of alpha * average_pooling(x) + (1 - alpha) * max_pooling(x) to put inside the network...

x = Conv2D(32, 3, activation='relu')(...)
a = AveragePooling2D()(x)
a = Lambda(lambda xx : xx*alpha)(a)
m = MaxPooling2D()(x)
m = Lambda(lambda xx : xx*(1-alpha))(m)
x = Add()([a,m])

with alpha = 0.xx in the range [0,1]

user_103750490
Related