# TCPα: Margin-Controlled Confidence estimation for reliable Music Information Retrieval

arXiv: https://arxiv.org/abs/2608.20326 · HTML: https://mlsanity.com/paper/2608.20326/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 29 of its 45 references that are in the map.
- First posted on arXiv 20 Aug 2026.

## Builds on (most cited first, 29 of 29)

- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning](https://mlsanity.com/paper/1506.02142/) (2015)
- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks](https://mlsanity.com/paper/1610.02136/) (2016)
- [Energy-based Out-of-distribution Detection](https://mlsanity.com/paper/2010.03759/) (2020)
- [Class-Balanced Loss Based on Effective Number of Samples](https://mlsanity.com/paper/1901.05555/) (2019)
- [Selective Classification for Deep Neural Networks](https://mlsanity.com/paper/1705.08500/) (2017)
- [Weight Uncertainty in Neural Networks](https://mlsanity.com/paper/1505.05424/) (2015)
- [Measuring Calibration in Deep Learning](https://mlsanity.com/paper/1904.01685/) (2019)
- [SelectiveNet: A Deep Neural Network with an Integrated Reject Option](https://mlsanity.com/paper/1901.09192/) (2019)
- [Addressing Failure Prediction by Learning Model Confidence](https://mlsanity.com/paper/1910.04851/) (2019)
- [Delving into Deep Imbalanced Regression](https://mlsanity.com/paper/2102.09554/) (2021)
- [Optimal strategies for reject option classifiers](https://mlsanity.com/paper/2101.12523/) (2021)
- [DOCTOR: A Simple Method for Detecting Misclassification Errors](https://mlsanity.com/paper/2106.02395/) (2021)
- [Rethinking Confidence Calibration for Failure Prediction](https://mlsanity.com/paper/2303.02970/) (2023)
- [Confidence Estimation via Auxiliary Models](https://mlsanity.com/paper/2012.06508/) (2020)
- [Weakly Labelled AudioSet Tagging with Attention Neural Networks](https://mlsanity.com/paper/1903.00765/) (2019)
- [30+ Years of Source Separation Research: Achievements and Future Challenges](https://mlsanity.com/paper/2501.11837/) (2025)
- [How Flawed Is ECE? An Analysis via Logit Smoothing](https://mlsanity.com/paper/2402.10046/) (2024)
- [Explainable Deep Learning Analysis for Raga Identification in Indian Art Music](https://mlsanity.com/paper/2406.02443/) (2024)
- [Recognizing Ornaments in Vocal Indian Art Music with Active Annotation](https://mlsanity.com/paper/2505.04419/) (2025)
- [Towards Efficient and Real-Time Piano Transcription Using Neural Autoregressive Models](https://mlsanity.com/paper/2404.06818/) (2024)
- [Identification and Clustering of Unseen Ragas in Indian Art Music](https://mlsanity.com/paper/2411.18611/) (2024)
- [The Inverse Drum Machine: Source Separation Through Joint Transcription and Analysis-by-Synthesis](https://mlsanity.com/paper/2505.03337/) (2025)
- [Uncertainty Calibration for Deep Audio Classifiers](https://mlsanity.com/paper/2206.13071/) (2022)
- [Interactive singing melody extraction based on active adaptation](https://mlsanity.com/paper/2402.07599/) (2024)
- [Meta-learning-based percussion transcription and tbarala identification from low-resource audio](https://mlsanity.com/paper/2501.04742/) (2025)
- [Improving Active Learning for Melody Estimation by Disentangling Uncertainties](https://mlsanity.com/paper/2509.17375/) (2025)
- [Snapping Matters: Context-Aware Onset Refinement for Automatic Music Transcription](https://mlsanity.com/paper/2606.11903/) (2026)

## Abstract (arXiv)

Deep neural networks are often overconfident, assigning high confidence even to incorrect predictions. Consequently, users lack a reliable signal for deciding when a prediction can be trusted. Post-hoc confidence estimation addresses this by training a lightweight auxiliary head over a frozen classifier. Existing targets, however, suffer from inherent ambiguity: they assign overlapping confidence values to correct and incorrect predictions, while errors near the decision boundary receive confidence scores indistinguishable from correct predictions. In this work, we propose TCPα, a novel confidence target that resolves these limitations by introducing a margin-controlled penalty for misclassified samples. We prove that TCPα guarantees complete separation between the target values of correct and incorrect predictions, with a separation margin that is independent of the number of classes and increases monotonically with the penalty parameter. Since accurate classifiers naturally produce very few errors, learning these targets results in a severely imbalanced regression problem. We therefore present a systematic study of training strategies for learning under this imbalance and identify an effective training configuration through extensive ablation studies. We evaluate the proposed approach on r=aga identification, investigate its robustness under domain shift, and further validate it on frame-wise ornamentation detection without modifying the selected configuration. Across all settings, TCPα consistently outperforms existing confidence targets for failure prediction. Rejecting only the least-confident 8% of predictions improves the base model's macro-F1 from 0.89 to 0.98, while fine-tuning the confidence head with only 5% labeled samples from a new corpus effectively restores performance under domain shift.