Owing to their remarkable learning (and relearning) capabilities, deep neural networks (DNNs) find use in numerous real-world applications. However, the learning of these data-driven machine learning models is generally as good as the data available to them for training. Hence, training datasets with long-tail distribution pose a challenge for DNNs, since the DNNs trained on them may provide a varying degree of classification performance across different output classes. While the overall bias of such networks is already highlighted in existing works, this work identifies the node bias that leads to a varying sensitivity of the nodes for different output classes. To the best of our knowledge, this is the first work highlighting this unique challenge in DNNs, discussing its probable causes, and providing open challenges for this new research direction. We support our reasoning using an empirical case study of the networks trained on a real-world dataset.

翻译：由于其卓越的学习（以及再学习）能力，深度神经网络被广泛应用于众多实际场景。然而，这些数据驱动型机器学习模型的学习效果通常取决于训练数据的质量。因此，具有长尾分布的训练数据集对深度神经网络构成了挑战——基于此类数据训练的模型在不同输出类别上的分类性能可能参差不齐。尽管现有研究已经揭示了这类网络的整体偏差，但本工作识别出导致节点对不同输出类别产生不同敏感性的节点偏差。据我们所知，这是首次聚焦深度神经网络中这一独特挑战的研究，探讨了其可能成因，并为这一新研究方向提出了开放性挑战。我们通过在真实数据集上训练网络形成的实证案例研究来支撑上述推论。