Auto Seg-Loss: Searching Metric Surrogates for Semantic Segmentation
Auto Seg-Loss: Searching Metric Surrogates for Semantic Segmentation
复制标题
自动分段丢失：搜索语义分段的度量替代
DOI：
--
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Jifeng Dai
中科院分区：
文献类型：
--
作者：
Hao Li;Chenxin Tao;Xizhou Zhu;Xiaogang Wang;Gao Huang;Jifeng Dai
设计合适的损失函数是训练深度网络的关键。特别是在语义分割领域，针对不同的场景提出了不同的评价指标。尽管广泛采用的交叉熵损失及其变体取得了成功，但损失函数和评估指标之间的不一致会降低网络性能。同时，手动设计每个特定度量的损失函数需要专业知识和大量人力。在本文中，我们提出通过搜索每个度量的可微代理损失来自动设计度量特定的损失函数。我们用参数化函数代替度量中的不可微操作，并进行参数搜索以优化损失曲面的形状。引入两个约束来正则化搜索空间，提高搜索效率。在PASCAL VOC和cityscape上进行的大量实验表明，搜索的代理损失始终优于手动设计的损失函数。搜索损失可以很好地推广到其他数据集和网络。应发布代码。
Designing proper loss functions is essential in training deep networks. Especially in the field of semantic segmentation, various evaluation metrics have been proposed for diverse scenarios. Despite the success of the widely adopted cross-entropy loss and its variants, the mis-alignment between the loss functions and evaluation metrics degrades the network performance. Meanwhile, manually designing loss functions for each specific metric requires expertise and significant manpower. In this paper, we propose to automate the design of metric-specific loss functions by searching differentiable surrogate losses for each metric. We substitute the non-differentiable operations in the metrics with parameterized functions, and conduct parameter search to optimize the shape of loss surfaces. Two constraints are introduced to regularize the search space and make the search efficient. Extensive experiments on PASCAL VOC and Cityscapes demonstrate that the searched surrogate losses outperform the manually designed loss functions consistently. The searched losses can generalize well to other datasets and networks. Code shall be released.
DOI：
--
发表时间：
2016-11
期刊：
ArXiv
影响因子：
--
作者：
Barret Zoph;Quoc V. Le
通讯作者：
Barret Zoph;Quoc V. Le