论文标题
深度音频分类器的不确定性校准
Uncertainty Calibration for Deep Audio Classifiers
论文作者
论文摘要
尽管深度神经网络（DNNS）在音频分类任务中取得了巨大的成功，但它们的不确定性校准仍未得到探索。当它确定其预测时，应进行良好的模型应该是准确的，并且表明何时可能不准确。在这项工作中，我们研究了深度音频分类器的不确定性校准。特别是，我们从经验上研究了流行校准方法的性能：（i）蒙特卡洛辍学方法，（ii）集合，（iii）局灶性损失和（iv）光谱分类高斯工艺（SNGP），在音频分类数据集上。为此，我们评估了（I-IV），以应对环境声音和音乐流派分类的任务。结果表明，未校准的深度音频分类器可能过于自信，并且SNGP在本文的两个数据集中表现最好，并且非常有效。
Although deep Neural Networks (DNNs) have achieved tremendous success in audio classification tasks, their uncertainty calibration are still under-explored. A well-calibrated model should be accurate when it is certain about its prediction and indicate high uncertainty when it is likely to be inaccurate. In this work, we investigate the uncertainty calibration for deep audio classifiers. In particular, we empirically study the performance of popular calibration methods: (i) Monte Carlo Dropout, (ii) ensemble, (iii) focal loss, and (iv) spectral-normalized Gaussian process (SNGP), on audio classification datasets. To this end, we evaluate (i-iv) for the tasks of environment sound and music genre classification. Results indicate that uncalibrated deep audio classifiers may be over-confident, and SNGP performs the best and is very efficient on the two datasets of this paper.
