为什么大多数在城市景观上都可以使用加权的横向凝胶？
重量跨entroy（WCE）有助于处理不平衡的数据集，而CityScapes非常不平衡如下所示：
如果我们检查最佳基准测试在此数据集中，大多数作品都使用Bare CE作为损失函数。如果有任何特殊原因会导致语义分割在 miou 评估中的任务任务，我就不会明白。
我特别问，因为我在一个更高的不平衡数据集中工作（以1：1000的比率为1：1000，当Bare CE在MIOU指标上胜过WCE时，我感到非常惊讶。
我发现WCE可以从少数阶级产生许多误报，但是除此之外，还有更多原因吗？
Weight Cross-Entroy (WCE) helps to handle an imbalanced dataset, and Cityscapes is quite imbalanced as seen below:
If we check the best benchmarks on this dataset, most of the works use bare CE as a loss function. I don't get it if there are any special causes that would lead WCE to a worse result for semantic segmentation tasks on the mIoU evaluation.
I'm especially asking because I'm working in an even higher unbalanced dataset (multi-minority classes on the ratio of 1:1000 to the majority classes) and got very surprised when bare CE outperformed WCE on the mIoU metric.
I found so far that WCE can yield many false positives from minority classes, but beyond that, would there be more reasons for it?
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
阅读 tsung-yi lin，Priya Goyal，Ross Girshick，Kaiming He和Piotrdrolár 密集对象检测的焦点损失 （ICCV 2017）。他们讨论了CE损失的缺点，而当课程不平衡并争论（非常令人信服的）时，WCE根本无法解决CE的限制。
CE损失永远不会达到零：即使预测是完美的，它始终具有非零梯度。 CE致力于增加不同类别之间的边缘。结果，当班级之间存在不平衡时，CE将对占主导地位的“更确定”以及在少数群体上犯下更少的错误而付出平等的努力。将权重放在CE上不会对此行为做出根本性的改变。
相比之下，在这种情况下，您实际想要的是忽略已经正确预测的样本，并努力纠正错误的预测。这通常是通过局灶性损失的硬性阴性开采来实现的。
Read Tsung-Yi Lin, Priya Goyal, Ross Girshick, Kaiming He and Piotr Dollár Focal Loss for Dense Object Detection (ICCV 2017). They discuss in length the shortcoming of CE loss when classes are unbalanced and argue (quite compellingly) that WCE simply does not address this limitation of CE.
CE loss never goes to zero: it always has non-zero gradients even if the prediction is perfect. CE strives to increase the margin between the different classes. As a result, when there is an imbalance between classes, CE will put an equal effort into being "more certain" about the dominant class as well as making fewer mistakes on the minority class. Putting weights on the CE will not make a fundamental change to this behavior.
In contrast, what you actually want from a loss function, in this case, is to ignore samples that you already predict correctly, and make an effort to correct wrong predictions. This is usually achieved via hard-negative mining of Focal loss.