NCTV: Neural Clamping Toolkit and Visualization for Neural Network Calibration
Lei Hsiung, Yung-Chen Tang, Pin-Yu Chen, Tsung-Yi Ho
Abstract
With the advancement of deep learning technology, neural networks have demonstrated their excellent ability to provide accurate predictions in many tasks. However, a lack of consideration for neural network calibration will not gain trust from humans, even for high-accuracy models. In this regard, the gap between the confidence of the model's predictions and the actual correctness likelihood must be bridged to derive a well-calibrated model. In this paper, we introduce the Neural Clamping Toolkit, the first open-source framework designed to help developers employ state-of-the-art model-agnostic calibrated models. Furthermore, we provide animations and interactive sections in the demonstration to familiarize researchers with calibration in neural networks. A Colab tutorial on utilizing our toolkit is also introduced.
BibTeX
@article{Hsiung_Tang_Chen_Ho_2024, title={NCTV: Neural Clamping Toolkit and Visualization for Neural Network Calibration}, volume={37}, url={https://ojs.aaai.org/index.php/AAAI/article/view/27074}, DOI={10.1609/aaai.v37i13.27074}, abstractNote={With the advancement of deep learning technology, neural networks have demonstrated their excellent ability to provide accurate predictions in many tasks. However, a lack of consideration for neural network calibration will not gain trust from humans, even for high-accuracy models. In this regard, the gap between the confidence of the model’s predictions and the actual correctness likelihood must be bridged to derive a well-calibrated model. In this paper, we introduce the Neural Clamping Toolkit, the first open-source framework designed to help developers employ state-of-the-art model-agnostic calibrated models. Furthermore, we provide animations and interactive sections in the demonstration to familiarize researchers with calibration in neural networks. A Colab tutorial on utilizing our toolkit is also introduced.}, number={13}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Hsiung, Lei and Tang, Yung-Chen and Chen, Pin-Yu and Ho, Tsung-Yi}, year={2024}, month={Jul.}, pages={16446-16448} }