在本文中，我們提出了一個分類信心度網路（CLCNet），它可以判斷分類模型是否正確 的分類了輸入樣本。其可使用任意維度的分類結果向量作為輸入，並返回一個信心度分 數作為輸出，該分數代表了樣本被正確分類的可能性。我們可以在由多個SOTA（stateof- the-art）分類模型組成的簡單級聯結構系統中使用CLCNet，並通過實驗表明該系統具 有以下優勢：1. 系統可以自定義推理每張照片的平均計算量（FLOPs）。2、在相同的 計算量下，系統的性能可以超過任何與系統中的模型結構相同但尺寸不同的模型。實 際上，這就是一種新型的集成學習方法。與一般的模型融合一樣，它可以達到比單一 分類模型更好的性能，但該系統只需要比一般的模型融合更少的計算量。代碼已經上傳 至Github：https://github.com/yaoching0/CLCNet-Rethinking-of-Ensemble-Modeling

In this paper, we propose a Classification Confidence Network (CLCNet) that can determine whether the classification model classifies input samples correctly. The proposed model can take a classification result in the form of vector in any dimension, and return a confidence score as output, which represents the probability of an instance being classified correctly. We can utilize CLCNet in a simple cascade structure system consisting of several SOTA (state-of-the-art) classification models, and our experiments show that the system can achieve the following advantages: 1. The system can customize the average computation requirement (FLOPs) per image while inference. 2. Under the same computation requirement, the performance of the system can exceed any model that has identical structure with the model in the system, but different in size. In fact, we consider our cascade structure system as a new type of ensemble modeling. Like general ensemble modeling, it can achieve higher performance than single classification model, yet our system requires much less computation than general ensemble modeling. The code have been uploaded to a github repository: https://github.com/yaoching0/CLCNet-Rethinking-of-Ensemble-Modeling.

[1] Li Yuan, Qibin Hou, Zihang Jiang, Jiashi Feng, and Shuicheng Yan. Volo: Vision outlooker for visual recognition. arXiv preprint arXiv:2106.13112, 2021.
[2] Zhuang Liu, Hanzi Mao, Chao-Yuan Wu, Christoph Feichtenhofer, Trevor Darrell, and Saining Xie. A convnet for the 2020s. arXiv preprint arXiv:2201.03545, 2022.
[3] Andy Brock, Soham De, Samuel L Smith, and Karen Simonyan. High-performance large-scale image recognition without normalization. In International Conference on Machine Learning, pages 1059–1071. PMLR, 2021.
[4] Mingxing Tan and Quoc Le. Efficientnetv2: Smaller models and faster training. In International Conference on Machine Learning, pages 10096–10106. PMLR, 2021.
[5] Mingxing Tan and Quoc Le. Efficientnet: Rethinking model scaling for convolutional neural networks. In International conference on machine learning, pages 6105–6114. PMLR, 2019.
[6] Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. Imagenet: A largescale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, pages 248–255. Ieee, 2009.
[7] Sheng Wan, Tung-Yu Wu, Wing H Wong, and Chen-Yi Lee. Confnet: predict with confidence. In 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 2921–2925. IEEE, 2018.
[8] Sercan O Arık and Tomas Pfister. Tabnet: Attentive interpretable tabular learning. In AAAI, volume 35, pages 6679–6687, 2021.
[9] Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems, 30, 2017.
[10] Sethu Vijayakumar and Stefan Schaal. Locally weighted projection regression: An o (n) algorithm for incremental real time learning in high dimensional space. In Proceedings of the seventeenth international conference on machine learning (ICML 2000), volume 1, pages 288–293. Morgan Kaufmann, 2000.
[11] Kaggle. Rossmann store sales. Accessed 10 November 2019. https://www.kaggle.com/c/rossmann-store-sales, 2015.
[12] Andre Martins and Ramon Astudillo. From softmax to sparsemax: A sparse model of attention and multi-label classification. In International conference on machine learning, pages 1614–1623. PMLR, 2016.
[13] Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.
[14] Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, et al. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929, 2020.
[15] Qizhe Xie, Minh-Thang Luong, Eduard Hovy, and Quoc V Le. Self-training with noisy student improves imagenet classification. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10687–10698, 2020.