论文标题
基于相关功能的基于辅助细胞，用于节能自然误差
Relevant-features based Auxiliary Cells for Energy Efficient Detection of Natural Errors
论文作者
论文摘要
深度神经网络已在许多分类任务上表现出最先进的表现。但是，他们没有固有的能力来识别他们的预测何时是错误的。最近，最近有几项努力来检测自然错误，但建议的机制提出了额外的能量需求。为了解决这个问题，我们提出了一个隐藏层的分类器集合，以实现自然错误的能源有效检测。特别是，我们附加了基于相关的辅助细胞（RAC），这些辅助细胞（RAC）是经过相关特征培训的特定二进制线性分类器。 RAC的共识用于检测自然错误。基于RAC的组合置信度，可以尽早终止分类，从而导致节能检测。我们证明了我们的技术对各种图像分类数据集的有效性，例如CIFAR-10，CIFAR-100和TININ-IMAGENET。
Deep neural networks have demonstrated state-of-the-art performance on many classification tasks. However, they have no inherent capability to recognize when their predictions are wrong. There have been several efforts in the recent past to detect natural errors but the suggested mechanisms pose additional energy requirements. To address this issue, we propose an ensemble of classifiers at hidden layers to enable energy efficient detection of natural errors. In particular, we append Relevant-features based Auxiliary Cells (RACs) which are class specific binary linear classifiers trained on relevant features. The consensus of RACs is used to detect natural errors. Based on combined confidence of RACs, classification can be terminated early, thereby resulting in energy efficient detection. We demonstrate the effectiveness of our technique on various image classification datasets such as CIFAR-10, CIFAR-100 and Tiny-ImageNet.
