A Comparison of Five Multiple Instance Learning Pooling Functions for Sound Event Detection with Weak Labeling
This work addresses sound event detection for audio analysis applications, but it is incremental as it focuses on comparing existing pooling functions rather than introducing a new method.
The paper tackled the problem of sound event detection with weak labeling by comparing five multiple instance learning pooling functions, finding that linear softmax pooling performed best and enabling a system called TALNet to achieve state-of-the-art audio tagging on Audio Set and strong localization on the DCASE 2017 challenge.
Sound event detection (SED) entails two subtasks: recognizing what types of sound events are present in an audio stream (audio tagging), and pinpointing their onset and offset times (localization). In the popular multiple instance learning (MIL) framework for SED with weak labeling, an important component is the pooling function. This paper compares five types of pooling functions both theoretically and experimentally, with special focus on their performance of localization. Although the attention pooling function is currently receiving the most attention, we find the linear softmax pooling function to perform the best among the five. Using this pooling function, we build a neural network called TALNet. It is the first system to reach state-of-the-art audio tagging performance on Audio Set, while exhibiting strong localization performance on the DCASE 2017 challenge at the same time.