连接深度神经网络中的概念凸性与人机对齐
Connecting Concept Convexity and Human-Machine Alignment in Deep Neural Networks
AI总结：
该研究基于行为数据分析预训练与微调 vision transformer，发现潜在表征凸性与人机对齐相关但关系复杂，为可解释、可靠 AI 提供新视角。
摘要：
理解神经网络如何与人类认知过程对齐，是开发更具可解释性和更可靠 AI 系统的关键一步。受人类认知理论启发，本研究基于行为数据，考察神经网络表征中的“凸性”与“人机对齐”之间的关系。我们在预训练和微调的 vision transformer 模型中发现这两个维度之间存在相关性。研究结果表明，神经网络潜在空间中形成的凸区域在一定程度上与人类定义的类别相对应，并反映人类在认知任务中使用的相似性关系。尽管针对对齐进行优化通常会增强凸性，但通过微调提高凸性对对齐产生的影响并不一致，这表明二者之间存在复杂关系。本研究为理解潜在表征凸性与人机对齐之间的关系迈出了第一步。
英文摘要：
Understanding how neural networks align with human cognitive processes is a crucial step toward developing more interpretable and reliable AI systems. Motivated by theories of human cognition, this study examines the relationship between \emph{convexity} in neural network representations and \emph{human-machine alignment} based on behavioral data. We identify a correlation between these two dimensions in pretrained and fine-tuned vision transformer models. Our findings suggest that the convex regions formed in latent spaces of neural networks to some extent align with human-defined categories and reflect the similarity relations humans use in cognitive tasks. While optimizing for alignment generally enhances convexity, increasing convexity through fine-tuning yields inconsistent effects on alignment, which suggests a complex relationship between the two. This study presents a first step toward understanding the relationship between the convexity of latent representations and human-machine alignment.
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。
