On calibration of modern neural networks
2017
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.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
33
References
1527
Citations
NaN
KQI