Adaptive pooling operators for weakly labeled sound event detection
B. Mcfee, J. Salamon, and J. P. Bello
IEEE/ACM Transactions on Audio, Speech and Language Processing, 26(11): 2180â2193, Nov. 2018.
[IEEE][PDF][BibTeX][Copyright][experiments][code release]
Using AutoPool
>pip install autopool
from autopool import AutoPool bag_pred = AutoPool(axis=1)(instance_pred)
Definition
Constrained and Regularized AutoPool
CAP with non-negative a:
bag_pred = AutoPool(axis=1, kernel_constraint=keras.constraints.non_neg())(instance_pred)
bag_pred = AutoPool(axis=1, kernel_constraint=keras.constraints.max_norm(alpha_max, axis=0))(instance_pred)
RAP with l2 regularized a:
bag_pred = AutoPool(axis=1, kernel_regularizer=keras.regularizers.l2(l=1e-4))(instance_pred)
Multi-label
Experimental Results
For more experimental results and a detailed evaluation of AutoPool compared to other pooling functions across several datasets, please see our paper.