随机格上的表示学习
Representation Learning on a Random Lattice
AI总结:
本文从几何视角将深度神经网络的特征视为学习坐标系,提出将通用数据分布建模为随机格,并利用渗透理论分析其性质,将特征分为上下文、组件和表面三类,为机制可解释性提供了理论支持。
摘要:
将深度神经网络学习到的表示分解为可解释特征,能极大提升其安全性与可靠性。为更好理解特征,我们采用几何视角,将其视为映射嵌入数据分布的学习坐标系。我们提出将通用数据分布建模为随机格,并利用渗透理论分析其性质。学习特征被分为上下文特征、组件特征和表面特征。该模型在定性上与机制可解释性的近期发现一致,并为未来研究指明了方向。
英文摘要:
Decomposing a deep neural network's learned representations into interpretable features could greatly enhance its safety and reliability. To better understand features, we adopt a geometric perspective, viewing them as a learned coordinate system for mapping an embedded data distribution. We motivate a model of a generic data distribution as a random lattice and analyze its properties using percolation theory. Learned features are categorized into context, component, and surface features. The model is qualitatively consistent with recent findings in mechanistic interpretability and suggests directions for future research.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。