# Constructing Composite Features for Interpretable Music-Tagging

arXiv: https://arxiv.org/abs/2603.28644 · HTML: https://mlsanity.com/paper/2603.28644/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 32 references that are in the map.
- First posted on arXiv 30 Mar 2026.

## Builds on (most cited first, 9 of 9)

- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [PANNs: Large-Scale Pretrained Audio Neural Networks for Audio Pattern Recognition](https://mlsanity.com/paper/1912.10211/) (2019)
- [musicnn: Pre-trained convolutional neural networks for music audio tagging](https://mlsanity.com/paper/1909.06654/) (2019)
- [End-to-end learning for music audio tagging at scale](https://mlsanity.com/paper/1711.02520/) (2017)
- [Semi-Supervised Music Tagging Transformer](https://mlsanity.com/paper/2111.13457/) (2021)
- [Masked Modeling Duo: Learning Representations by Encouraging Both Networks to Model the Input](https://mlsanity.com/paper/2210.14648/) (2022)
- [Omnizart: A General Toolbox for Automatic Music Transcription](https://mlsanity.com/paper/2106.00497/) (2021)
- [Perceptual Musical Features for Interpretable Audio Tagging](https://mlsanity.com/paper/2312.11234/) (2023)
- [Towards Unified Music Emotion Recognition across Dimensional and Categorical Models](https://mlsanity.com/paper/2502.03979/) (2025)

## Abstract (arXiv)

Combining multiple audio features can improve the performance of music tagging, but common deep learning-based feature fusion methods often lack interpretability. To address this problem, we propose a Genetic Programming (GP) pipeline that automatically evolves composite features by mathematically combining base music features, thereby capturing synergistic interactions while preserving interpretability. This approach provides representational benefits similar to deep feature fusion without sacrificing interpretability. Experiments on the MTG-Jamendo and GTZAN datasets demonstrate consistent improvements compared to state-of-the-art systems across base feature sets at different abstraction levels. It should be noted that most of the performance gains are noticed within the first few hundred GP evaluations, indicating that effective feature combinations can be identified under modest search budgets. The top evolved expressions include linear, nonlinear, and conditional forms, with various low-complexity solutions at top performance aligned with parsimony pressure to prefer simpler expressions. Analyzing these composite features further reveals which interactions and transformations tend to be beneficial for tagging, offering insights that remain opaque in black-box deep models.