Sentence examples for pooling algorithm from inspiring English sources

Zeiler et al. [20] have proposed a stochastic pooling algorithm which randomly picks the activation within the pooling region based on a multinomial distribution.

Fig. 4 Pooling algorithm event handlers.

As was mentioned earlier, Fig. 4 illustrates the Pooling algorithm.

Since we make adjustments according to the sparsity degree of feature maps in the adaptive pooling algorithm, the experimental results further confirm that whether a pooling method is suitable depends on the sparsity degree of features activations.

Additionally, we also tested the performance of the novel stochastic pooling algorithm proposed in [20] which has been introduced in the 'Related work' section and made a comparison with conventional pooling methods and adaptive pooling we proposed.

Classification accuracy under each condition is presented in Table 1 and Figure 7, where Avg denotes average pooling, Max denotes max pooling, and Ada denotes the adaptive pooling algorithm proposed in 'Pooling operation' section.

Although several novel pooling algorithms such as stochastic pooling [20] have been proposed, average pooling and max pooling are still the most commonly used approaches.

It is also interesting to note that the unbalanced pooling algorithms returned very balanced partitions.

We start by noting that for both data sets the results obtained by the balanced pooling algorithms turned out to be almost identical to those obtained by the unbalanced pooling algorithms for each of the two cost functions we used.

It includes the following components: an objective-feature_label table, a dataset pool, an algorithm pool, feature_label discovery function, data_attribute discovery function, data_fusion function, and algorithm_fusion function.

For extracting similarity information from coefficients, the alignment-pooling algorithm is applied to obtain pooled features.