# Improving Post-Processing of Audio Event Detectors Using Reinforcement Learning

arXiv: https://arxiv.org/abs/2208.09201 · HTML: https://mlsanity.com/paper/2208.09201/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 20 references that are in the map.
- First posted on arXiv 19 Aug 2022.

## Builds on (most cited first, 10 of 10)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Asynchronous Methods for Deep Reinforcement Learning](https://mlsanity.com/paper/1602.01783/) (2016)
- [On the Properties of Neural Machine Translation: Encoder-Decoder Approaches](https://mlsanity.com/paper/1409.1259/) (2014)
- [LIFT: Reinforcement Learning in Computer Systems by Learning From Demonstrations](https://mlsanity.com/paper/1808.07903/) (2018)
- [A Deep Reinforcement Learning Approach to Audio-Based Navigation in a Multi-Speaker Environment](https://mlsanity.com/paper/2105.04488/) (2021)
- [Do Autonomous Agents Benefit from Hearing?](https://mlsanity.com/paper/1905.04192/) (2019)
- [Evaluation of post-processing algorithms for polyphonic sound event detection](https://mlsanity.com/paper/1906.06909/) (2019)
- [Agents that Listen: High-Throughput Reinforcement Learning with Multiple Sensory Systems](https://mlsanity.com/paper/2107.02195/) (2021)
- [A Deep Reinforcement Learning Blind AI in DareFightingICE](https://mlsanity.com/paper/2205.07444/) (2022)

## Built on by (most cited first, 1 of 1)

- [NEUROSEC: FPGA-Based Neuromorphic Audio Security](https://mlsanity.com/paper/2401.12055/) (2024)

## Abstract (arXiv)

We apply post-processing to the class probability distribution outputs of audio event classification models and employ reinforcement learning to jointly discover the optimal parameters for various stages of a post-processing stack, such as the classification thresholds and the kernel sizes of median filtering algorithms used to smooth out model predictions. To achieve this we define a reinforcement learning environment where: 1) a state is the class probability distribution provided by the model for a given audio sample, 2) an action is the choice of a candidate optimal value for each parameter of the post-processing stack, 3) the reward is based on the classification accuracy metric we aim to optimize, which is the audio event-based macro F1-score in our case. We apply our post-processing to the class probability distribution outputs of two audio event classification models submitted to the DCASE Task4 2020 challenge. We find that by using reinforcement learning to discover the optimal per-class parameters for the post-processing stack that is applied to the outputs of audio event classification models, we can improve the audio event-based macro F1-score (the main metric used in the DCASE challenge to compare audio event classification accuracy) by 4-5% compared to using the same post-processing stack with manually tuned parameters.