通过在模型内部合并相似卷积核，提升医学图像分类模型的鲁棒性。
In-Model Merging for Enhancing the Robustness of Medical Imaging Classification Models
- 训练时动态合并深层相似卷积核，增强模型抗干扰能力。
- 在4个数据集上显著超越传统训练模型，提升明显。
- 适用于各类CNN架构，特别适合对可靠性要求高的医学领域。
模型融合是一种有效提升模型性能的策略，相较于集成学习更高效，因推理时无需额外计算。然而，现有研究极少探讨是否可在单一模型内部进行融合以增强其鲁棒性，而这一特性在医学图像领域尤为关键。本文首次提出一种名为InMerge的新方法，通过在训练过程中选择性地合并单个卷积神经网络（CNN）深层中的相似卷积核，来增强模型的鲁棒性。我们还从理论上揭示了影响融合效果的关键特征，为社区提供重要参考。实验表明，该方法在4个主流数据集上对多种CNN架构均具可行性与有效性，经InMerge训练的模型显著优于常规训练模型。代码将公开。
原文摘要 · Abstract (English)
Model merging is an effective strategy to merge multiple models for enhancing model performances, and more efficient than ensemble learning as it will not introduce extra computation into inference. However, limited research explores if the merging process can occur within one model and enhance the model's robustness, which is particularly critical in the medical image domain. In the paper, we are the first to propose in-model merging (InMerge), a novel approach that enhances the model's robustness by selectively merging similar convolutional kernels in the deep layers of a single convolutional neural network (CNN) during the training process for classification. We also analytically reveal important characteristics that affect how in-model merging should be performed, serving as an insightful reference for the community. We demonstrate the feasibility and effectiveness of this technique for different CNN architectures on 4 prevalent datasets. The proposed InMerge-trained model surpasses the typically-trained model by a substantial margin. The code will be made public.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。