arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1706.04599cs.LG
收藏

现代神经网络的校正

On Calibration of Modern Neural Networks

Cornell University(康奈尔大学)

Chuan Guo, Geoff Pleiss, Yu Sun, Kilian Q. Weinberger

摘要:

置信度校准-预测代表真实正确性可能性的概率估计的问题-对于许多应用中的分类模型很重要。我们发现,现代神经网络与十年前的网络不同,校准得很差。通过大量的实验,我们观察到,深度,宽度,权重衰减,和批量归一化是影响校准的重要因素。我们评估了各种后处理校准方法在最先进的图像和文档分类数据集架构上的性能。我们的分析和实验不仅提供了对神经网络学习的见解,而且还为实际设置提供了简单而直接的方法:在大多数数据集上,温度缩放-普拉特缩放的单参数变体-在校准预测方面令人惊讶地有效。

英文摘要:

Confidence calibration -- the problem of predicting probability estimates representative of the true correctness likelihood -- is important for classification models in many applications. We discover that modern neural networks, unlike those from a decade ago, are poorly calibrated. Through extensive experiments, we observe that depth, width, weight decay, and Batch Normalization are important factors influencing calibration. We evaluate the performance of various post-processing calibration methods on state-of-the-art architectures with image and document classification datasets. Our analysis and experiments not only offer insights into neural network learning, but also provide a simple and straightforward recipe for practical settings: on most datasets, temperature scaling -- a single-parameter variant of Platt Scaling -- is surprisingly effective at calibrating predictions.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑