On Calibration of Modern Neural Networks
Confidence calibration – the problem of predicting probability estimates representative of the true correctness likelihood – is important for classification models in many applications. We discover...
On Calibration of Modern Neural Networks Chuan Guo, Geoff Pleiss, Yu Sun, Kilian Q. Weinberger Proceedings of the 34th International Conference on Machine Learning , PMLR 70:1321-1330, 2017. Abstract 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 facto
Explore this link on the map →related reading
- [1706.04599] On Calibration of Modern Neural Networksarxiv.org
- On Getting Confidence Estimates from Neural Networks | Bharath's notesbharathpbhat.github.io
- Bayesian Neural Networkscs.toronto.edu
- [2003.07892] Calibration of Pre-trained Transformersarxiv.org
- Zoom In: An Introduction to Circuitsdistill.pub
- Neural networks and deep learningneuralnetworksanddeeplearning.com
- Aman's AI Journal • Primers • Ilya Sutskever's Top 30aman.ai
- The Decade of Deep Learning | Leo Gaobmk.sh
- A Recipe for Training Neural Networkskarpathy.github.io
- On neural scaling and the quanta hypothesisericjmichaud.com
- Home - colah's blogcolah.github.io
- arxiv.org/pdf/1805.08522arxiv.org