# Adaptive pooling operators for weakly labeled sound event detection

arXiv: https://arxiv.org/abs/1804.10070 · HTML: https://mlsanity.com/paper/1804.10070/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Apr 2018.

## Built on by (most cited first, 22 of 22)

- [Contrastive Positive Sample Propagation along the Audio-Visual Event Line](https://mlsanity.com/paper/2211.09980/) (2022)
- [Connecting the Dots between Audio and Text without Parallel Data through Visual Knowledge Transfer](https://mlsanity.com/paper/2112.08995/) (2021)
- [Sound Event Detection Transformer: An Event-based End-to-End Model for Sound Event Detection](https://mlsanity.com/paper/2110.02011/) (2021)
- [Comparison of Methods Generalizing Max- and Average-Pooling](https://mlsanity.com/paper/2103.01746/) (2021)
- [Self-Supervision Enhances Instance-based Multiple Instance Learning Methods in Digital Pathology: A Benchmark Study](https://mlsanity.com/paper/2505.01109/) (2025)
- [Distilling Image Classifiers in Object Detectors](https://mlsanity.com/paper/2106.05209/) (2021)
- [Improving weakly supervised sound event detection with self-supervised auxiliary tasks](https://mlsanity.com/paper/2106.06858/) (2021)
- [The impact of non-target events in synthetic soundscapes for sound event detection](https://mlsanity.com/paper/2109.14061/) (2021)
- [Sound Event Detection in Urban Audio With Single and Multi-Rate PCEN](https://mlsanity.com/paper/2102.03468/) (2021)
- [A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets](https://mlsanity.com/paper/2107.07927/) (2021)
- [Elementary, My Dear Watson: Non-Invasive Neural Keyword Spotting in the LibriBrain Dataset](https://mlsanity.com/paper/2510.21038/) (2025)
- [Joint framework with deep feature distillation and adaptive focal loss for weakly supervised audio tagging and acoustic event detection](https://mlsanity.com/paper/2103.12388/) (2021)
- [Writing in The Air: Unconstrained Text Recognition from Finger Movement Using Spatio-Temporal Convolution](https://mlsanity.com/paper/2104.09021/) (2021)
- [Multitask vocal burst modeling with ResNets and pre-trained paralinguistic Conformers](https://mlsanity.com/paper/2206.12494/) (2022)
- [On the choice of the optimal temporal support for audio classification with Pre-trained embeddings](https://mlsanity.com/paper/2312.14005/) (2023)
- [The Accuracy Cost of Weakness: A Theoretical Analysis of Fixed-Segment Weak Labeling for Events in Time](https://mlsanity.com/paper/2502.09363/) (2025)
- [Attention Mamba: Time Series Modeling with Adaptive Pooling Acceleration and Receptive Field Enhancements](https://mlsanity.com/paper/2504.02013/) (2025)
- [Improving Bird Classification with Primary Color Additives](https://mlsanity.com/paper/2507.18334/) (2025)
- [QASTAnet: A DNN-based Quality Metric for Spatial Audio](https://mlsanity.com/paper/2509.16715/) (2025)
- [WS-IMUBench: Can Weakly Supervised Methods from Audio, Image, and Video Be Adapted for IMU-based Temporal Action Localization?](https://mlsanity.com/paper/2602.01850/) (2026)
- [Profy: Interpretable Visualization of Expertise-Dependent Motor Skills Toward Supporting Piano Practice](https://mlsanity.com/paper/2606.10627/) (2026)
- [DiFA: Dual Evidence Fusion and Aggregation for Token-Level Text Anomaly Detection](https://mlsanity.com/paper/2609.22136/) (2026)

## Abstract (arXiv)

Sound event detection (SED) methods are tasked with labeling segments of audio recordings by the presence of active sound sources. SED is typically posed as a supervised machine learning problem, requiring strong annotations for the presence or absence of each sound source at every time instant within the recording. However, strong annotations of this type are both labor- and cost-intensive for human annotators to produce, which limits the practical scalability of SED methods. In this work, we treat SED as a multiple instance learning (MIL) problem, where training labels are static over a short excerpt, indicating the presence or absence of sound sources but not their temporal locality. The models, however, must still produce temporally dynamic predictions, which must be aggregated (pooled) when comparing against static labels during training. To facilitate this aggregation, we develop a family of adaptive pooling operators---referred to as auto-pool---which smoothly interpolate between common pooling operators, such as min-, max-, or average-pooling, and automatically adapt to the characteristics of the sound sources in question. We evaluate the proposed pooling operators on three datasets, and demonstrate that in each case, the proposed methods outperform non-adaptive pooling operators for static prediction, and nearly match the performance of models trained with strong, dynamic annotations. The proposed method is evaluated in conjunction with convolutional neural networks, but can be readily applied to any differentiable model for time-series label prediction.