损失不变投影作为学习表示的无源探针

中文摘要 AI 辅助

神经网络中学习到的特征表示通常包含超出最终任务输出直接使用的结构之外的结构。我们使用\u201c无源探针\u201d来研究这种结构，这些探针在训练过程中对表示应用固定的、未经训练的、与属性无关的投影。我们通过$S^2$上的预测任务来激发这种方法，其中等价的向量和厄米参数化揭示了一个额外的损失不变迹坐标。这激发了一种通用构造，其中固定的随机投影作为学习特征的观察者。由于观察者是损失不变的且独立于所研究的属性，可达性的变化反映的是表示相对于固定观察者的变化，而非观察者自身的适应。我们表明，无源探针的集成可以直接反映任务相关信息，如目标对齐。在我们的构造下，最终难度的可达性在不同任务中演化不同。在表面法线估计和图像修复的回归任务中，它在训练期间增加，而在图像分类中则保持接近其初始水平。与学习线性探针的比较进一步表明，可恢复性和无源可达性在训练期间可以不同地演化。总之，这些结果展示了无源探针如何分别表征表示几何的变化和最终任务难度的可达性。

英文摘要

Learned feature representations in neural networks often contain structure beyond that directly used by the final task output. We study this structure using $\textit{passive probes}$ that apply fixed, untrained, property-independent projections to representations as they evolve during training. We motivate this approach through the task of prediction on $S^2$ where equivalent vector and Hermitian parameterizations reveal an additional loss-invariant trace coordinate. This motivates a general construction in which fixed random projections serve as observers of learned features. Because the observer is loss-invariant and independent of the property being studied, changes in accessibility reflect changes in the representation relative to the fixed observer rather than adaptation of the observer itself. We show that ensembles of passive probes can directly reflect task-relevant information such as target alignment. Under our constructions, the accessibility of eventual difficulty evolves differently across tasks. It increases during training in the regression tasks of surface-normal estimation and image inpainting but remains near its initial level in image classification. Comparisons with learned linear probes further show that recoverability and passive accessibility can evolve differently during training. Together, these results show how passive probes can separately characterize changes in representation geometry and the accessibility of eventual task difficulty.

发表机构

• Linköping University（林雪平大学）

机构由 AI 辅助整理，请以论文原文为准。