IEEE Trans Pattern Anal Mach Intell. 2022 Oct;44(10):6823-6838. doi: 10.1109/TPAMI.2021.3094625. Epub 2022 Sep 14.
One of the most prominent attributes of Neural Networks (NNs) constitutes their capability of learning to extract robust and descriptive features from high dimensional data, like images. Hence, such an ability renders their exploitation as feature extractors particularly frequent in an abundance of modern reasoning systems. Their application scope mainly includes complex cascade tasks, like multi-modal recognition and deep Reinforcement Learning (RL). However, NNs induce implicit biases that are difficult to avoid or to deal with and are not met in traditional image descriptors. Moreover, the lack of knowledge for describing the intra-layer properties -and thus their general behavior- restricts the further applicability of the extracted features. With the paper at hand, a novel way of visualizing and understanding the vector space before the NNs' output layer is presented, aiming to enlighten the deep feature vectors' properties under classification tasks. Main attention is paid to the nature of overfitting in the feature space and its adverse effect on further exploitation. We present the findings that can be derived from our model's formulation and we evaluate them on realistic recognition scenarios, proving its prominence by improving the obtained results.
神经网络（NNs）最显著的特征之一是它们能够从高维数据（如图像）中学习提取健壮和描述性的特征。因此，这种能力使得它们作为特征提取器在大量现代推理系统中得到了广泛应用。它们的应用范围主要包括复杂的级联任务，如多模态识别和深度强化学习（RL）。然而，NNs 会引入难以避免或处理的隐式偏差，这些偏差在传统的图像描述符中是不存在的。此外，缺乏描述层内特性（因此也是它们的一般行为）的知识限制了提取特征的进一步适用性。本文提出了一种在 NNs 输出层之前可视化和理解向量空间的新方法，旨在阐明分类任务下深特征向量的特性。主要关注特征空间中的过拟合性质及其对进一步开发的不利影响。我们从模型的公式中得出了可以得出的结论，并在现实的识别场景中对其进行了评估，通过提高获得的结果证明了其优越性。
