当前位置: X-MOL 学术Neural Process Lett. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Output Layer Multiplication for Class Imbalance Problem in Convolutional Neural Networks
Neural Processing Letters ( IF 2.6 ) Pub Date : 2020-10-19 , DOI: 10.1007/s11063-020-10366-w
Zhao Yang , Yuanxin Zhu , Tie Liu , Sai Zhao , Yunyan Wang , Dapeng Tao

Convolutional neural networks (CNNs) have demonstrated remarkable performance in the field of computer vision. However, they are prone to suffer from the class imbalance problem, in which the number of some classes is significantly higher or lower than that of other classes. Commonly, there are two main strategies to handle the problem, including dataset-level methods via resampling and algorithmic-level methods by modifying the existing learning frameworks. However, most of these methods need extra data resampling or elaborate algorithm design. In this work we provide an effective but extremely simple approach to tackle the imbalance problem in CNNs with cross-entropy loss. Specifically, we multiply a coefficient \( \alpha > 1 \) to output of the last layer in a CNN model. With this modification, the final loss function can dynamically adjust the contributions of examples from different classes during the imbalanced training procedure. Because of its simplicity, the proposed method can be easily applied in the off-the-shelf models with little change. To prove the effectiveness on imbalance problem, we design three experiments on classification tasks of increasing complexity. The experimental results show that our approach could improve the convergence rate in the training stage and/or increase accuracy for test.



中文翻译:

卷积神经网络中类不平衡问题的输出层乘法

卷积神经网络(CNN)在计算机视觉领域表现出了卓越的性能。但是,他们容易遭受阶级失衡的问题,其中某些阶级的人数明显高于或低于其他阶级的人数。通常,有两种主要策略可以解决该问题,包括通过重采样进行数据集级别的方法以及通过修改现有学习框架进行算法级别的方法。但是,大多数这些方法都需要额外的数据重采样或精心设计的算法。在这项工作中,我们提供了一种有效但极其简单的方法来解决带有交叉熵损失的CNN中的不平衡问题。具体来说,我们将系数\(\ alpha> 1 \)相乘到CNN模型中最后一层的输出。通过这种修改,最终损失函数可以在不平衡训练过程中动态调整不同类别的示例的贡献。由于其简单性,所提出的方法几乎无需更改即可轻松应用于现有模型。为了证明在不平衡问题上的有效性,我们设计了三个实验,以提高复杂度。实验结果表明,我们的方法可以提高训练阶段的收敛速度和/或提高测试的准确性。

更新日期:2020-10-20
down
wechat
bug